model-notebook/modelNotebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# League of Legends eSports Predictor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hello!\n",
    "\n",
    "http://www.lolesports-predictor.com/\n",
    "\n",
    "### Introduction\n",
    "\n",
    "This notebook is my personal project to predict a winner between two LoL esports teams. At first I wanted to try to predict a professional game's winner based on champion selection, only to come to a dissapointing conclusion that it is almost impossible, resulting close to a 50-50 guess.\n",
    "\n",
    "I then decided to move my goal to simply predict games based on who's playing who and on which side of the map. I gathered all my data from [gol.gg](https://gol.gg/esports/home/) and preprocessed it. I tried a bunch of different algorithms and models, evaluated them and came to a conclusion which of them is the best. I then created a demo section to predict your own games.\n",
    "\n",
    "### About the model\n",
    "\n",
    "The model was taught with all the regular season games from 2020 to February 2022 in *LEC*, *LCS* and *LCK*, as well as some *LPL* (all the LPL data between 2020-2022 was not usable) with a total amount of 1477 games. The model would most likely benefit from adding more games to reach atleast 2000 games.\n",
    "\n",
    "The predictor uses team strenghts as an average of their whole splits and does not take into account variables such as individual players statistics, teams head-to-head records, or team's current situation (such as an ongoing win or losing streaks). For these reasons the model will also make more reliable predictions towards the end of the season. It does not know if any sudden roster swaps or other outside of the game events occur.\n",
    "\n",
    "I managed to get a good 70% accuracy and 80% recall without overfitting, which is around the same what an average active watcher and some professional analysts get, based on [IWDominate's](https://www.twitch.tv/iwilldominate) and his viewers predictions.\n",
    "\n",
    "### Project aims\n",
    "\n",
    "The aim of this project was not only to predict the winners but also strenghten my own data preprocessing and machine learning skills with Python, as well as learning how to deploy and manage apps with Heroku and Flask.\n",
    "\n",
    "I do not recommend to use the model for betting, but if you decided to do so let me know how you did!\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section I import all the necessary libraries and read the data from the HTML files. I managed to read the data straight from URL only after I had done this, so for now I don't bother to change them to that. Local files for the model training might not be a bad thing, since now it can be rerun every time independently from the gol.gg website status. However the predictor uses URLs because it needs to be up to date with all the team's strenghts for real time predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n",
      "tensorflow version: 2.7.0\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "import time\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras import models, layers, losses, metrics, optimizers\n",
    "from sklearn.naive_bayes import GaussianNB, MultinomialNB\n",
    "from sklearn.metrics import accuracy_score\n",
    "import pydotplus\n",
    "import graphviz\n",
    "from sklearn import linear_model\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn import tree\n",
    "import requests\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = (20,10)\n",
    "print(f'tensorflow version: {tf.__version__}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we gather the teams and match histories from the files. It's not pretty looking and could be divided into it's own python file, but I will leave it here to show a thorough progress of my work."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Game</th>\n",
       "      <th>Unnamed: 1</th>\n",
       "      <th>Score</th>\n",
       "      <th>Unnamed: 3</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "      <th>Patch</th>\n",
       "      <th>Date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>VIT vs Fnatic</td>\n",
       "      <td>Team Vitality</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>G2 Esports vs Astralis</td>\n",
       "      <td>G2 Esports</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>Astralis</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>SK Gaming vs MAD Lions</td>\n",
       "      <td>SK Gaming</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>MAD Lions</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>Rogue vs MSF</td>\n",
       "      <td>Rogue</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>Misfits Gaming</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>Team BDS vs XL</td>\n",
       "      <td>Excel Esports</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>Team BDS</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>Team BDS vs Fnatic</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>Team BDS</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>MSF vs Astralis</td>\n",
       "      <td>Astralis</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>Misfits Gaming</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>G2 Esports vs XL</td>\n",
       "      <td>Excel Esports</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>G2 Esports</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>SK Gaming vs Rogue</td>\n",
       "      <td>Rogue</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>SK Gaming</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>VIT vs MAD Lions</td>\n",
       "      <td>MAD Lions</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>Team Vitality</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.1</td>\n",
       "      <td>2022-01-14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Game     Unnamed: 1  Score      Unnamed: 3 Unnamed: 4  \\\n",
       "45           VIT vs Fnatic  Team Vitality  0 - 1          Fnatic      WEEK1   \n",
       "46  G2 Esports vs Astralis     G2 Esports  1 - 0        Astralis      WEEK1   \n",
       "47  SK Gaming vs MAD Lions      SK Gaming  1 - 0       MAD Lions      WEEK1   \n",
       "48            Rogue vs MSF          Rogue  1 - 0  Misfits Gaming      WEEK1   \n",
       "49          Team BDS vs XL  Excel Esports  0 - 1        Team BDS      WEEK1   \n",
       "50      Team BDS vs Fnatic         Fnatic  1 - 0        Team BDS      WEEK1   \n",
       "51         MSF vs Astralis       Astralis  0 - 1  Misfits Gaming      WEEK1   \n",
       "52        G2 Esports vs XL  Excel Esports  0 - 1      G2 Esports      WEEK1   \n",
       "53      SK Gaming vs Rogue          Rogue  1 - 0       SK Gaming      WEEK1   \n",
       "54        VIT vs MAD Lions      MAD Lions  1 - 0   Team Vitality      WEEK1   \n",
       "\n",
       "    Patch        Date  \n",
       "45   12.1  2022-01-15  \n",
       "46   12.1  2022-01-15  \n",
       "47   12.1  2022-01-15  \n",
       "48   12.1  2022-01-15  \n",
       "49   12.1  2022-01-15  \n",
       "50   12.1  2022-01-14  \n",
       "51   12.1  2022-01-14  \n",
       "52   12.1  2022-01-14  \n",
       "53   12.1  2022-01-14  \n",
       "54   12.1  2022-01-14  "
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#-------------------------// 2022 //---------------------------------#\n",
    "\n",
    "        # -------------- // Spring // -------------- #\n",
    "\n",
    "                # ------ // History // ------ #\n",
    "\n",
    "lck2022SpringHistory = pd.read_html(\"data/LCK2022SpringHistory.html\")\n",
    "lck2022SpringHistory = pd.concat(lck2022SpringHistory)\n",
    "\n",
    "lec2022SpringHistory = pd.read_html(\"data/LEC2022SpringHistory.html\")\n",
    "lec2022SpringHistory = pd.concat(lec2022SpringHistory)\n",
    "\n",
    "lcs2022SpringHistory = pd.read_html(\"data/LCS2022SpringHistory.html\")\n",
    "lcs2022SpringHistory = pd.concat(lcs2022SpringHistory)\n",
    "\n",
    "\n",
    "                # ------ // Teams // ------ #\n",
    "\n",
    "lck2022SpringTeams = pd.read_html(\"data/LCK2022SpringTeams.html\")\n",
    "lck2022SpringTeams = pd.concat(lck2022SpringTeams)\n",
    "\n",
    "lec2022SpringTeams = pd.read_html(\"data/LEC2022SpringTeams.html\")\n",
    "lec2022SpringTeams = pd.concat(lec2022SpringTeams)\n",
    "\n",
    "lcs2022SpringTeams = pd.read_html(\"data/LCS2022SpringTeams.html\")\n",
    "lcs2022SpringTeams = pd.concat(lcs2022SpringTeams)\n",
    "\n",
    "#-------------------------// 2021 //---------------------------------#\n",
    "\n",
    "        # -------------- // Summer // -------------- #\n",
    "\n",
    "                # ------ // Teams // ------ #\n",
    "\n",
    "\n",
    "lec2021SummerTeams = pd.read_html(\"data/LEC2021SummerTeams.html\")\n",
    "lec2021SummerTeams = pd.concat(lec2021SummerTeams)\n",
    "\n",
    "lcs2021SummerTeams = pd.read_html(\"data/LCS2021SummerTeams.html\")\n",
    "lcs2021SummerTeams = pd.concat(lcs2021SummerTeams)\n",
    "\n",
    "lck2021SummerTeams = pd.read_html(\"data/LCK2021SummerTeams.html\")\n",
    "lck2021SummerTeams = pd.concat(lck2021SummerTeams)\n",
    "\n",
    "        # -------------- // History // -------------- #\n",
    "\n",
    "lcs2021SummerHistory = pd.read_html(\"data/LCS2021SummerHistory.html\")\n",
    "lcs2021SummerHistory = pd.concat(lcs2021SummerHistory)\n",
    "\n",
    "lec2021SummerHistory = pd.read_html(\"data/LEC2021SummerHistory.html\")\n",
    "lec2021SummerHistory = pd.concat(lec2021SummerHistory)\n",
    "\n",
    "lck2021SummerHistory = pd.read_html(\"data/LCK2021SummerHistory.html\")\n",
    "lck2021SummerHistory = pd.concat(lck2021SummerHistory)\n",
    "\n",
    "        # -------------- // Spring // -------------- #\n",
    "\n",
    "                # ------ // Teams // ------ #\n",
    "\n",
    "lec2021SpringTeams = pd.read_html(\"data/LEC2021SpringTeams.html\")\n",
    "lec2021SpringTeams = pd.concat(lec2021SpringTeams)\n",
    "\n",
    "lcs2021SpringTeams = pd.read_html(\"data/LCS2021SpringTeams.html\")\n",
    "lcs2021SpringTeams = pd.concat(lcs2021SpringTeams)\n",
    "\n",
    "lck2021SpringTeams = pd.read_html(\"data/LCK2021SpringTeams.html\")\n",
    "lck2021SpringTeams = pd.concat(lck2021SpringTeams)\n",
    "\n",
    "lpl2021SpringTeams = pd.read_html(\"data/LPL2021SpringTeams.html\")\n",
    "lpl2021SpringTeams = pd.concat(lpl2021SpringTeams)\n",
    "\n",
    "                # ------ // History // ------ #\n",
    "\n",
    "lec2021SpringHistory = pd.read_html(\"data/LEC2021SpringHistory.html\")\n",
    "lec2021SpringHistory = pd.concat(lec2021SpringHistory)\n",
    "\n",
    "lcs2021SpringHistory = pd.read_html(\"data/LCS2021SpringHistory.html\")\n",
    "lcs2021SpringHistory = pd.concat(lcs2021SpringHistory)\n",
    "\n",
    "lck2021SpringHistory = pd.read_html(\"data/LCK2021SpringHistory.html\")\n",
    "lck2021SpringHistory = pd.concat(lck2021SpringHistory)\n",
    "\n",
    "lpl2021SpringHistory = pd.read_html(\"data/LPL2021SpringHistory.html\")\n",
    "lpl2021SpringHistory = pd.concat(lpl2021SpringHistory)\n",
    "\n",
    "#-------------------------// 2020 //---------------------------------#\n",
    "\n",
    "        # -------------- // Summer // -------------- #\n",
    "\n",
    "                # ------ // Teams // ------ #\n",
    "\n",
    "lec2020SummerTeams = pd.read_html(\"data/LEC2020SummerTeams.html\")\n",
    "lec2020SummerTeams = pd.concat(lec2020SummerTeams)\n",
    "\n",
    "lcs2020SummerTeams = pd.read_html(\"data/LCS2020SummerTeams.html\")\n",
    "lcs2020SummerTeams = pd.concat(lcs2020SummerTeams)\n",
    "\n",
    "lck2020SummerTeams = pd.read_html(\"data/LCK2020SummerTeams.html\")\n",
    "lck2020SummerTeams = pd.concat(lck2020SummerTeams)\n",
    "\n",
    "lpl2020SummerTeams = pd.read_html(\"data/LPL2020SummerTeams.html\")\n",
    "lpl2020SummerTeams = pd.concat(lpl2020SummerTeams)\n",
    "\n",
    "                # ------ // History // ------ #\n",
    "\n",
    "lcs2020SummerHistory = pd.read_html(\"data/LCS2020SummerHistory.html\")\n",
    "lcs2020SummerHistory = pd.concat(lcs2020SummerHistory)\n",
    "\n",
    "lec2020SummerHistory = pd.read_html(\"data/LEC2020SummerHistory.html\")\n",
    "lec2020SummerHistory = pd.concat(lec2020SummerHistory)\n",
    "\n",
    "lck2020SummerHistory = pd.read_html(\"data/LCK2020SummerHistory.html\")\n",
    "lck2020SummerHistory = pd.concat(lck2020SummerHistory)\n",
    "\n",
    "lpl2020SummerHistory = pd.read_html(\"data/LPL2020SummerHistory.html\")\n",
    "lpl2020SummerHistory = pd.concat(lpl2020SummerHistory)\n",
    "\n",
    "        # -------------- // Spring // -------------- #\n",
    "\n",
    "                # ------ // Teams // ------ #\n",
    "\n",
    "lec2020SpringTeams = pd.read_html(\"data/LEC2020SpringTeams.html\")\n",
    "lec2020SpringTeams = pd.concat(lec2020SpringTeams)\n",
    "\n",
    "lcs2020SpringTeams = pd.read_html(\"data/LCS2020SpringTeams.html\")\n",
    "lcs2020SpringTeams = pd.concat(lcs2020SpringTeams)\n",
    "\n",
    "lck2020SpringTeams = pd.read_html(\"data/LCK2020SpringTeams.html\")\n",
    "lck2020SpringTeams = pd.concat(lck2020SpringTeams)\n",
    "\n",
    "                # ------ // History // ------ #\n",
    "\n",
    "lec2020SpringHistory = pd.read_html(\"data/LEC2020SpringHistory.html\")\n",
    "lec2020SpringHistory = pd.concat(lec2020SpringHistory)\n",
    "\n",
    "lcs2020SpringHistory = pd.read_html(\"data/LCS2020SpringHistory.html\")\n",
    "lcs2020SpringHistory = pd.concat(lcs2020SpringHistory)\n",
    "\n",
    "lck2020SpringHistory = pd.read_html(\"data/LCK2020SpringHistory.html\")\n",
    "lck2020SpringHistory = pd.concat(lck2020SpringHistory)\n",
    "\n",
    "lec2022SpringHistory.tail(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is what one splits match history looks like (10 games)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CSM</th>\n",
       "      <th>DPM</th>\n",
       "      <th>DRA%</th>\n",
       "      <th>DRA@15</th>\n",
       "      <th>DRAPG</th>\n",
       "      <th>Deaths / game</th>\n",
       "      <th>FB%</th>\n",
       "      <th>FT%</th>\n",
       "      <th>GD@15</th>\n",
       "      <th>GDM</th>\n",
       "      <th>...</th>\n",
       "      <th>Name</th>\n",
       "      <th>Region</th>\n",
       "      <th>Season</th>\n",
       "      <th>TD@15</th>\n",
       "      <th>Towers killed</th>\n",
       "      <th>Towers lost</th>\n",
       "      <th>VWPM</th>\n",
       "      <th>WCPM</th>\n",
       "      <th>WPM</th>\n",
       "      <th>Win rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>31.6</td>\n",
       "      <td>1670.0</td>\n",
       "      <td>31.3</td>\n",
       "      <td>0.33</td>\n",
       "      <td>1.56</td>\n",
       "      <td>15.4</td>\n",
       "      <td>55.6</td>\n",
       "      <td>22.2</td>\n",
       "      <td>-1089.0</td>\n",
       "      <td>-264.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Astralis</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>-0.4</td>\n",
       "      <td>2.9</td>\n",
       "      <td>9.1</td>\n",
       "      <td>1.56</td>\n",
       "      <td>1.36</td>\n",
       "      <td>3.4</td>\n",
       "      <td>0.0%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>32.1</td>\n",
       "      <td>1903.0</td>\n",
       "      <td>48.9</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.44</td>\n",
       "      <td>12.8</td>\n",
       "      <td>66.7</td>\n",
       "      <td>44.4</td>\n",
       "      <td>-60.0</td>\n",
       "      <td>-62.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Excel Esports</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>-0.2</td>\n",
       "      <td>6.9</td>\n",
       "      <td>6.6</td>\n",
       "      <td>1.49</td>\n",
       "      <td>1.39</td>\n",
       "      <td>3.6</td>\n",
       "      <td>44.4%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33.7</td>\n",
       "      <td>1981.0</td>\n",
       "      <td>66.3</td>\n",
       "      <td>1.00</td>\n",
       "      <td>3.11</td>\n",
       "      <td>11.7</td>\n",
       "      <td>33.3</td>\n",
       "      <td>77.8</td>\n",
       "      <td>1208.0</td>\n",
       "      <td>74.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>32.1</td>\n",
       "      <td>1855.0</td>\n",
       "      <td>52.4</td>\n",
       "      <td>0.67</td>\n",
       "      <td>2.56</td>\n",
       "      <td>11.6</td>\n",
       "      <td>55.6</td>\n",
       "      <td>44.4</td>\n",
       "      <td>68.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>...</td>\n",
       "      <td>G2 Esports</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.2</td>\n",
       "      <td>7.0</td>\n",
       "      <td>6.2</td>\n",
       "      <td>1.27</td>\n",
       "      <td>1.60</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>32.8</td>\n",
       "      <td>2064.0</td>\n",
       "      <td>43.7</td>\n",
       "      <td>0.22</td>\n",
       "      <td>2.11</td>\n",
       "      <td>16.1</td>\n",
       "      <td>66.7</td>\n",
       "      <td>66.7</td>\n",
       "      <td>397.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>...</td>\n",
       "      <td>MAD Lions</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.1</td>\n",
       "      <td>7.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.33</td>\n",
       "      <td>1.46</td>\n",
       "      <td>3.3</td>\n",
       "      <td>55.6%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>33.2</td>\n",
       "      <td>2039.0</td>\n",
       "      <td>51.3</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.22</td>\n",
       "      <td>13.6</td>\n",
       "      <td>44.4</td>\n",
       "      <td>55.6</td>\n",
       "      <td>-244.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Misfits Gaming</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.1</td>\n",
       "      <td>7.8</td>\n",
       "      <td>4.9</td>\n",
       "      <td>1.30</td>\n",
       "      <td>1.38</td>\n",
       "      <td>3.1</td>\n",
       "      <td>55.6%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>32.9</td>\n",
       "      <td>2279.0</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.44</td>\n",
       "      <td>6.7</td>\n",
       "      <td>66.7</td>\n",
       "      <td>77.8</td>\n",
       "      <td>1707.0</td>\n",
       "      <td>376.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Rogue</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>9.4</td>\n",
       "      <td>2.6</td>\n",
       "      <td>1.16</td>\n",
       "      <td>1.46</td>\n",
       "      <td>3.2</td>\n",
       "      <td>100.0%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>32.0</td>\n",
       "      <td>1785.0</td>\n",
       "      <td>47.2</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.22</td>\n",
       "      <td>15.8</td>\n",
       "      <td>44.4</td>\n",
       "      <td>22.2</td>\n",
       "      <td>-2006.0</td>\n",
       "      <td>-232.0</td>\n",
       "      <td>...</td>\n",
       "      <td>SK Gaming</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>-0.6</td>\n",
       "      <td>3.7</td>\n",
       "      <td>8.9</td>\n",
       "      <td>1.31</td>\n",
       "      <td>1.31</td>\n",
       "      <td>3.3</td>\n",
       "      <td>22.2%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>32.0</td>\n",
       "      <td>1712.0</td>\n",
       "      <td>44.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.11</td>\n",
       "      <td>14.6</td>\n",
       "      <td>55.6</td>\n",
       "      <td>44.4</td>\n",
       "      <td>-602.0</td>\n",
       "      <td>-134.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Team BDS</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>-0.3</td>\n",
       "      <td>4.8</td>\n",
       "      <td>7.7</td>\n",
       "      <td>1.45</td>\n",
       "      <td>1.25</td>\n",
       "      <td>3.3</td>\n",
       "      <td>33.3%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>34.9</td>\n",
       "      <td>2041.0</td>\n",
       "      <td>47.6</td>\n",
       "      <td>1.11</td>\n",
       "      <td>2.11</td>\n",
       "      <td>13.2</td>\n",
       "      <td>11.1</td>\n",
       "      <td>44.4</td>\n",
       "      <td>620.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Team Vitality</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.8</td>\n",
       "      <td>5.8</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.46</td>\n",
       "      <td>3.1</td>\n",
       "      <td>55.6%</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    CSM     DPM  DRA%  DRA@15  DRAPG  Deaths / game   FB%   FT%   GD@15  \\\n",
       "0  31.6  1670.0  31.3    0.33   1.56           15.4  55.6  22.2 -1089.0   \n",
       "1  32.1  1903.0  48.9    0.89   2.44           12.8  66.7  44.4   -60.0   \n",
       "2  33.7  1981.0  66.3    1.00   3.11           11.7  33.3  77.8  1208.0   \n",
       "3  32.1  1855.0  52.4    0.67   2.56           11.6  55.6  44.4    68.0   \n",
       "4  32.8  2064.0  43.7    0.22   2.11           16.1  66.7  66.7   397.0   \n",
       "5  33.2  2039.0  51.3    0.89   2.22           13.6  44.4  55.6  -244.0   \n",
       "6  32.9  2279.0  66.7    0.89   2.44            6.7  66.7  77.8  1707.0   \n",
       "7  32.0  1785.0  47.2    0.89   2.22           15.8  44.4  22.2 -2006.0   \n",
       "8  32.0  1712.0  44.7    0.89   2.11           14.6  55.6  44.4  -602.0   \n",
       "9  34.9  2041.0  47.6    1.11   2.11           13.2  11.1  44.4   620.0   \n",
       "\n",
       "     GDM  ...            Name Region  Season  TD@15  Towers killed  \\\n",
       "0 -264.0  ...        Astralis    EUW     S12   -0.4            2.9   \n",
       "1  -62.0  ...   Excel Esports    EUW     S12   -0.2            6.9   \n",
       "2   74.0  ...          Fnatic    EUW     S12    0.6            6.9   \n",
       "3   19.0  ...      G2 Esports    EUW     S12    0.2            7.0   \n",
       "4   33.0  ...       MAD Lions    EUW     S12    0.1            7.1   \n",
       "5   90.0  ...  Misfits Gaming    EUW     S12    0.1            7.8   \n",
       "6  376.0  ...           Rogue    EUW     S12    0.6            9.4   \n",
       "7 -232.0  ...       SK Gaming    EUW     S12   -0.6            3.7   \n",
       "8 -134.0  ...        Team BDS    EUW     S12   -0.3            4.8   \n",
       "9  101.0  ...   Team Vitality    EUW     S12    0.0            6.8   \n",
       "\n",
       "   Towers lost  VWPM  WCPM  WPM Win rate  \n",
       "0          9.1  1.56  1.36  3.4     0.0%  \n",
       "1          6.6  1.49  1.39  3.6    44.4%  \n",
       "2          5.6  1.20  1.44  3.4    66.7%  \n",
       "3          6.2  1.27  1.60  3.4    66.7%  \n",
       "4          6.0  1.33  1.46  3.3    55.6%  \n",
       "5          4.9  1.30  1.38  3.1    55.6%  \n",
       "6          2.6  1.16  1.46  3.2   100.0%  \n",
       "7          8.9  1.31  1.31  3.3    22.2%  \n",
       "8          7.7  1.45  1.25  3.3    33.3%  \n",
       "9          5.8  1.20  1.46  3.1    55.6%  \n",
       "\n",
       "[10 rows x 29 columns]"
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "demo = lec2022SpringTeams\n",
    "demo = demo.drop(columns=[0, 1])\n",
    "demo = demo.dropna()\n",
    "demo.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And this is what teams strenghts look like in one split. For each team it has the following data:\n",
    "\n",
    "|Variable|Meaning|\n",
    "|--|--|\n",
    "|CSM|Creep score per minute|\n",
    "|DPM|Damage per minute|\n",
    "|DRA%|Percentage of dragons killed|\n",
    "|DRA@15|Dragons at 15|\n",
    "|Deaths / game|Deaths of all player per game| \n",
    "|FB%|First blood percentage|\n",
    "|FT%|First turret percentage|\n",
    "|GD@15|Gold difference @ 15|\n",
    "|GDM|Gold difference per minute|\n",
    "|GPM|Gold per minute|\n",
    "|Game duration|Game duration|\n",
    "|Games|Number of games|\n",
    "|HER%|Percentage of heralds killed| \n",
    "|HERPG|Heralds per game|\n",
    "|K:D|Average Kills/Deaths per game|\n",
    "|Kills / game|Kills of all player per game|\n",
    "|NASH%|Percentage of Barons killed|\n",
    "|NASHPG|Barons per game|\n",
    "|Name|Name of the team|\n",
    "|Region|Region of the team|\n",
    "|Season|Season| \n",
    "|TD@15|Tower differential @ 15|\n",
    "|Towers killed|Towers killed per game|\n",
    "|Towers lost|Towers lost per game|\n",
    "|VWPM|Vision wards per minute|\n",
    "|WCPM|Wards cleared per minute|\n",
    "|WPM|Wards per minute|\n",
    "|Win rate|Win rate of the team|\n",
    "\n",
    "Next I will merge the match histories and the team strenghts into a one Dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Game</th>\n",
       "      <th>Unnamed: 1</th>\n",
       "      <th>Score</th>\n",
       "      <th>Unnamed: 3</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "      <th>Patch</th>\n",
       "      <th>Date</th>\n",
       "      <th>CSM_x</th>\n",
       "      <th>DPM_x</th>\n",
       "      <th>DRA%_x</th>\n",
       "      <th>...</th>\n",
       "      <th>Name_y</th>\n",
       "      <th>Region_y</th>\n",
       "      <th>Season_y</th>\n",
       "      <th>TD@15_y</th>\n",
       "      <th>Towers killed_y</th>\n",
       "      <th>Towers lost_y</th>\n",
       "      <th>VWPM_y</th>\n",
       "      <th>WCPM_y</th>\n",
       "      <th>WPM_y</th>\n",
       "      <th>Win rate_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>VIT vs Fnatic</td>\n",
       "      <td>Team Vitality</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.10</td>\n",
       "      <td>2022-01-15</td>\n",
       "      <td>34.9</td>\n",
       "      <td>2041.0</td>\n",
       "      <td>47.6</td>\n",
       "      <td>...</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SK Gaming vs Fnatic</td>\n",
       "      <td>SK Gaming</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>WEEK1</td>\n",
       "      <td>12.10</td>\n",
       "      <td>2022-01-16</td>\n",
       "      <td>32.0</td>\n",
       "      <td>1785.0</td>\n",
       "      <td>47.2</td>\n",
       "      <td>...</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Rogue vs Fnatic</td>\n",
       "      <td>Rogue</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>WEEK3</td>\n",
       "      <td>12.20</td>\n",
       "      <td>2022-01-28</td>\n",
       "      <td>32.9</td>\n",
       "      <td>2279.0</td>\n",
       "      <td>66.7</td>\n",
       "      <td>...</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>MAD Lions vs Fnatic</td>\n",
       "      <td>MAD Lions</td>\n",
       "      <td>0 - 1</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>WEEK4</td>\n",
       "      <td>12.20</td>\n",
       "      <td>2022-02-04</td>\n",
       "      <td>32.8</td>\n",
       "      <td>2064.0</td>\n",
       "      <td>43.7</td>\n",
       "      <td>...</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Fnatic vs XL</td>\n",
       "      <td>Excel Esports</td>\n",
       "      <td>1 - 0</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>WEEK4</td>\n",
       "      <td>12.20</td>\n",
       "      <td>2022-02-05</td>\n",
       "      <td>32.1</td>\n",
       "      <td>1903.0</td>\n",
       "      <td>48.9</td>\n",
       "      <td>...</td>\n",
       "      <td>Fnatic</td>\n",
       "      <td>EUW</td>\n",
       "      <td>S12</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>JD Gaming vs FPX</td>\n",
       "      <td>JD Gaming</td>\n",
       "      <td>2 - 0</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>WEEK7</td>\n",
       "      <td>10.14</td>\n",
       "      <td>2020-07-19</td>\n",
       "      <td>33.5</td>\n",
       "      <td>2013.0</td>\n",
       "      <td>54.6</td>\n",
       "      <td>...</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>CN</td>\n",
       "      <td>S10</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>OMG vs FPX</td>\n",
       "      <td>OMG</td>\n",
       "      <td>0 - 2</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>WEEK8</td>\n",
       "      <td>10.14</td>\n",
       "      <td>2020-07-21</td>\n",
       "      <td>32.2</td>\n",
       "      <td>1831.0</td>\n",
       "      <td>46.3</td>\n",
       "      <td>...</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>CN</td>\n",
       "      <td>S10</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>EDG vs FPX</td>\n",
       "      <td>Edward Gaming</td>\n",
       "      <td>1 - 2</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>WEEK8</td>\n",
       "      <td>10.14</td>\n",
       "      <td>2020-07-25</td>\n",
       "      <td>31.7</td>\n",
       "      <td>1797.0</td>\n",
       "      <td>47.6</td>\n",
       "      <td>...</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>CN</td>\n",
       "      <td>S10</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>RNG vs FPX</td>\n",
       "      <td>Royal Never Give Up</td>\n",
       "      <td>1 - 2</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>WEEK5</td>\n",
       "      <td>10.13</td>\n",
       "      <td>2020-07-05</td>\n",
       "      <td>32.4</td>\n",
       "      <td>1803.0</td>\n",
       "      <td>52.4</td>\n",
       "      <td>...</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>CN</td>\n",
       "      <td>S10</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>Vici Gaming vs FPX</td>\n",
       "      <td>Vici Gaming</td>\n",
       "      <td>2 - 1</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>WEEK6</td>\n",
       "      <td>10.13</td>\n",
       "      <td>2020-07-09</td>\n",
       "      <td>32.8</td>\n",
       "      <td>1867.0</td>\n",
       "      <td>50.6</td>\n",
       "      <td>...</td>\n",
       "      <td>Funplus Phoenix</td>\n",
       "      <td>CN</td>\n",
       "      <td>S10</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7%</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1477 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Game           Unnamed: 1  Score       Unnamed: 3  \\\n",
       "1          VIT vs Fnatic        Team Vitality  0 - 1           Fnatic   \n",
       "3    SK Gaming vs Fnatic            SK Gaming  0 - 1           Fnatic   \n",
       "4        Rogue vs Fnatic                Rogue  1 - 0           Fnatic   \n",
       "5    MAD Lions vs Fnatic            MAD Lions  0 - 1           Fnatic   \n",
       "6           Fnatic vs XL        Excel Esports  1 - 0           Fnatic   \n",
       "..                   ...                  ...    ...              ...   \n",
       "131     JD Gaming vs FPX            JD Gaming  2 - 0  Funplus Phoenix   \n",
       "132           OMG vs FPX                  OMG  0 - 2  Funplus Phoenix   \n",
       "133           EDG vs FPX        Edward Gaming  1 - 2  Funplus Phoenix   \n",
       "134           RNG vs FPX  Royal Never Give Up  1 - 2  Funplus Phoenix   \n",
       "135   Vici Gaming vs FPX          Vici Gaming  2 - 1  Funplus Phoenix   \n",
       "\n",
       "    Unnamed: 4  Patch        Date  CSM_x   DPM_x  DRA%_x  ...  \\\n",
       "1        WEEK1  12.10  2022-01-15   34.9  2041.0    47.6  ...   \n",
       "3        WEEK1  12.10  2022-01-16   32.0  1785.0    47.2  ...   \n",
       "4        WEEK3  12.20  2022-01-28   32.9  2279.0    66.7  ...   \n",
       "5        WEEK4  12.20  2022-02-04   32.8  2064.0    43.7  ...   \n",
       "6        WEEK4  12.20  2022-02-05   32.1  1903.0    48.9  ...   \n",
       "..         ...    ...         ...    ...     ...     ...  ...   \n",
       "131      WEEK7  10.14  2020-07-19   33.5  2013.0    54.6  ...   \n",
       "132      WEEK8  10.14  2020-07-21   32.2  1831.0    46.3  ...   \n",
       "133      WEEK8  10.14  2020-07-25   31.7  1797.0    47.6  ...   \n",
       "134      WEEK5  10.13  2020-07-05   32.4  1803.0    52.4  ...   \n",
       "135      WEEK6  10.13  2020-07-09   32.8  1867.0    50.6  ...   \n",
       "\n",
       "              Name_y  Region_y  Season_y  TD@15_y Towers killed_y  \\\n",
       "1             Fnatic       EUW       S12      0.6             6.9   \n",
       "3             Fnatic       EUW       S12      0.6             6.9   \n",
       "4             Fnatic       EUW       S12      0.6             6.9   \n",
       "5             Fnatic       EUW       S12      0.6             6.9   \n",
       "6             Fnatic       EUW       S12      0.6             6.9   \n",
       "..               ...       ...       ...      ...             ...   \n",
       "131  Funplus Phoenix        CN       S10      0.0             6.1   \n",
       "132  Funplus Phoenix        CN       S10      0.0             6.1   \n",
       "133  Funplus Phoenix        CN       S10      0.0             6.1   \n",
       "134  Funplus Phoenix        CN       S10      0.0             6.1   \n",
       "135  Funplus Phoenix        CN       S10      0.0             6.1   \n",
       "\n",
       "     Towers lost_y  VWPM_y  WCPM_y WPM_y  Win rate_y  \n",
       "1              5.6    1.20    1.44   3.4       66.7%  \n",
       "3              5.6    1.20    1.44   3.4       66.7%  \n",
       "4              5.6    1.20    1.44   3.4       66.7%  \n",
       "5              5.6    1.20    1.44   3.4       66.7%  \n",
       "6              5.6    1.20    1.44   3.4       66.7%  \n",
       "..             ...     ...     ...   ...         ...  \n",
       "131            6.0    1.48    1.51   3.4       53.7%  \n",
       "132            6.0    1.48    1.51   3.4       53.7%  \n",
       "133            6.0    1.48    1.51   3.4       53.7%  \n",
       "134            6.0    1.48    1.51   3.4       53.7%  \n",
       "135            6.0    1.48    1.51   3.4       53.7%  \n",
       "\n",
       "[1477 rows x 65 columns]"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lec2022Spring = pd.merge(lec2022SpringHistory, lec2022SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lec2022Spring = pd.merge(lec2022Spring, lec2022SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lec2022 = lec2022Spring\n",
    "\n",
    "lck2022Spring = pd.merge(lck2022SpringHistory, lck2022SpringTeams, left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lck2022Spring = pd.merge(lck2022Spring, lck2022SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lck2022 = lck2022Spring\n",
    "\n",
    "lcs2022Spring = pd.merge(lcs2022SpringHistory, lcs2022SpringTeams, left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lcs2022Spring = pd.merge(lcs2022Spring, lcs2022SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lcs2022 = lcs2022Spring\n",
    "\n",
    "lec2021Spring = pd.merge(lec2021SpringHistory, lec2021SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lec2021Spring = pd.merge(lec2021Spring, lec2021SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lec2021Summer = pd.merge(lec2021SummerHistory, lec2021SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lec2021Summer = pd.merge(lec2021Summer, lec2021SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lec2021 = pd.concat([lec2021Summer, lec2021Spring])\n",
    "\n",
    "lck2021Spring = pd.merge(lck2021SpringHistory, lck2021SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lck2021Spring = pd.merge(lck2021Spring, lck2021SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lck2021Summer = pd.merge(lck2021SummerHistory, lck2021SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lck2021Summer = pd.merge(lck2021Summer, lck2021SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lck2021 = pd.concat([lck2021Summer, lck2021Spring])\n",
    "\n",
    "lcs2021Spring = pd.merge(lcs2021SpringHistory, lcs2021SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lcs2021Spring = pd.merge(lcs2021Spring, lcs2021SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lcs2021Summer = pd.merge(lcs2021SummerHistory, lcs2021SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lcs2021Summer = pd.merge(lcs2021Summer, lcs2021SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lcs2021 = pd.concat([lcs2021Summer, lcs2021Spring])\n",
    "\n",
    "lpl2021Spring = pd.merge(lpl2021SpringHistory, lpl2021SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lpl2021Spring = pd.merge(lpl2021Spring, lpl2021SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lpl2021 = pd.concat([lpl2021Spring])\n",
    "\n",
    "# --------------------- // 2020 // ----------------------- #\n",
    "\n",
    "lec2020Spring = pd.merge(lec2020SpringHistory, lec2020SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lec2020Spring = pd.merge(lec2020Spring, lec2020SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lec2020Summer = pd.merge(lec2020SummerHistory, lec2020SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lec2020Summer = pd.merge(lec2020Summer, lec2020SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lec2020 = pd.concat([lec2020Summer, lec2020Spring])\n",
    "\n",
    "lck2020Spring = pd.merge(lck2020SpringHistory, lck2020SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lck2020Spring = pd.merge(lck2020Spring, lck2020SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lck2020Summer = pd.merge(lck2020SummerHistory, lck2020SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lck2020Summer = pd.merge(lck2020Summer, lck2020SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lck2020 = pd.concat([lck2020Summer, lck2020Spring])\n",
    "\n",
    "lcs2020Spring = pd.merge(lcs2020SpringHistory, lcs2020SpringTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lcs2020Spring = pd.merge(lcs2020Spring, lcs2020SpringTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lcs2020Summer = pd.merge(lcs2020SummerHistory, lcs2020SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lcs2020Summer = pd.merge(lcs2020Summer, lcs2020SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lcs2020 = pd.concat([lcs2020Summer, lcs2020Spring])\n",
    "\n",
    "lpl2020Summer = pd.merge(lpl2020SummerHistory, lpl2020SummerTeams,  left_on=\"Unnamed: 1\", right_on=\"Name\")\n",
    "lpl2020Summer = pd.merge(lpl2020Summer, lpl2020SummerTeams, left_on=\"Unnamed: 3\", right_on=\"Name\")\n",
    "lpl2020 = pd.concat([lpl2020Summer])\n",
    "\n",
    "final = pd.concat([lec2022, lck2022, lec2021, lck2021, lcs2021, lpl2021, lec2020, lck2020, lcs2020, lpl2020])\n",
    "final = final[final[\"Score\"].str[0] != '-']\n",
    "final = final.drop(columns=['0_y', '0_x', '1_x', '1_y', '2_x', '2_y', '3_x', '3_y'])\n",
    "final.to_csv(\"data.csv\", index=False)\n",
    "final"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And this is what the merged dataframe looks like. It still needs some preprocessing, which I will do in the next section."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section I'm going to process the data into a model fittable format. I use a helper function to format the time from type <code>String</code> minutes and seconds into a <code>float</code> number seconds. I do the same kind of thing for score, formatting the <code>String</code> score results into a binary win/loss."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_sec(time_str):\n",
    "    m, s = time_str.split(':')\n",
    "    return float(60 * int(m) + int(s))\n",
    "\n",
    "def formatScore(score):\n",
    "    if score[0] == '0' or score == '1 - 2':\n",
    "        return [0, 1]\n",
    "    else:\n",
    "        return [1, 0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next I will handle all the columns that might have inappropriate values. These are for example <code>FT%</code> and <code>FB%</code>, which gol.gg displays as \"-\" marks if their values are zeros. I use the <code>get_sec(time_str)</code> function to format the time and <code>formatScore(score)</code> to format the scores to a binary wins or losses. I take <code>Strings</code> out of Seasons and Win rates. Lastly I drop all the columns that are not needed in the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Score</th>\n",
       "      <th>CSM_x</th>\n",
       "      <th>DPM_x</th>\n",
       "      <th>DRA%_x</th>\n",
       "      <th>DRA@15_x</th>\n",
       "      <th>DRAPG_x</th>\n",
       "      <th>Deaths / game_x</th>\n",
       "      <th>FB%_x</th>\n",
       "      <th>FT%_x</th>\n",
       "      <th>GD@15_x</th>\n",
       "      <th>...</th>\n",
       "      <th>NASH%_y</th>\n",
       "      <th>NASHPG_y</th>\n",
       "      <th>Season_y</th>\n",
       "      <th>TD@15_y</th>\n",
       "      <th>Towers killed_y</th>\n",
       "      <th>Towers lost_y</th>\n",
       "      <th>VWPM_y</th>\n",
       "      <th>WCPM_y</th>\n",
       "      <th>WPM_y</th>\n",
       "      <th>Win rate_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[0, 1]</td>\n",
       "      <td>34.9</td>\n",
       "      <td>2041.0</td>\n",
       "      <td>47.6</td>\n",
       "      <td>1.11</td>\n",
       "      <td>2.11</td>\n",
       "      <td>13.2</td>\n",
       "      <td>11.1</td>\n",
       "      <td>44.4</td>\n",
       "      <td>620.0</td>\n",
       "      <td>...</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[0, 1]</td>\n",
       "      <td>32.0</td>\n",
       "      <td>1785.0</td>\n",
       "      <td>47.2</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.22</td>\n",
       "      <td>15.8</td>\n",
       "      <td>44.4</td>\n",
       "      <td>22.2</td>\n",
       "      <td>-2006.0</td>\n",
       "      <td>...</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[1, 0]</td>\n",
       "      <td>32.9</td>\n",
       "      <td>2279.0</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.44</td>\n",
       "      <td>6.7</td>\n",
       "      <td>66.7</td>\n",
       "      <td>77.8</td>\n",
       "      <td>1707.0</td>\n",
       "      <td>...</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[0, 1]</td>\n",
       "      <td>32.8</td>\n",
       "      <td>2064.0</td>\n",
       "      <td>43.7</td>\n",
       "      <td>0.22</td>\n",
       "      <td>2.11</td>\n",
       "      <td>16.1</td>\n",
       "      <td>66.7</td>\n",
       "      <td>66.7</td>\n",
       "      <td>397.0</td>\n",
       "      <td>...</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>[1, 0]</td>\n",
       "      <td>32.1</td>\n",
       "      <td>1903.0</td>\n",
       "      <td>48.9</td>\n",
       "      <td>0.89</td>\n",
       "      <td>2.44</td>\n",
       "      <td>12.8</td>\n",
       "      <td>66.7</td>\n",
       "      <td>44.4</td>\n",
       "      <td>-60.0</td>\n",
       "      <td>...</td>\n",
       "      <td>66.7</td>\n",
       "      <td>0.89</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>6.9</td>\n",
       "      <td>5.6</td>\n",
       "      <td>1.20</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.4</td>\n",
       "      <td>66.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>[1, 0]</td>\n",
       "      <td>33.5</td>\n",
       "      <td>2013.0</td>\n",
       "      <td>54.6</td>\n",
       "      <td>0.72</td>\n",
       "      <td>2.44</td>\n",
       "      <td>10.7</td>\n",
       "      <td>46.2</td>\n",
       "      <td>41.0</td>\n",
       "      <td>245.0</td>\n",
       "      <td>...</td>\n",
       "      <td>45.1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>[0, 1]</td>\n",
       "      <td>32.2</td>\n",
       "      <td>1831.0</td>\n",
       "      <td>46.3</td>\n",
       "      <td>0.85</td>\n",
       "      <td>2.22</td>\n",
       "      <td>15.2</td>\n",
       "      <td>48.8</td>\n",
       "      <td>29.3</td>\n",
       "      <td>-619.0</td>\n",
       "      <td>...</td>\n",
       "      <td>45.1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>[0, 1]</td>\n",
       "      <td>31.7</td>\n",
       "      <td>1797.0</td>\n",
       "      <td>47.6</td>\n",
       "      <td>0.84</td>\n",
       "      <td>2.47</td>\n",
       "      <td>11.7</td>\n",
       "      <td>57.9</td>\n",
       "      <td>50.0</td>\n",
       "      <td>143.0</td>\n",
       "      <td>...</td>\n",
       "      <td>45.1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>[0, 1]</td>\n",
       "      <td>32.4</td>\n",
       "      <td>1803.0</td>\n",
       "      <td>52.4</td>\n",
       "      <td>1.05</td>\n",
       "      <td>2.41</td>\n",
       "      <td>11.4</td>\n",
       "      <td>64.1</td>\n",
       "      <td>48.7</td>\n",
       "      <td>927.0</td>\n",
       "      <td>...</td>\n",
       "      <td>45.1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>[1, 0]</td>\n",
       "      <td>32.8</td>\n",
       "      <td>1867.0</td>\n",
       "      <td>50.6</td>\n",
       "      <td>0.98</td>\n",
       "      <td>2.46</td>\n",
       "      <td>12.1</td>\n",
       "      <td>43.9</td>\n",
       "      <td>58.5</td>\n",
       "      <td>565.0</td>\n",
       "      <td>...</td>\n",
       "      <td>45.1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.48</td>\n",
       "      <td>1.51</td>\n",
       "      <td>3.4</td>\n",
       "      <td>53.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1477 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Score  CSM_x   DPM_x  DRA%_x  DRA@15_x  DRAPG_x  Deaths / game_x  FB%_x  \\\n",
       "1    [0, 1]   34.9  2041.0    47.6      1.11     2.11             13.2   11.1   \n",
       "3    [0, 1]   32.0  1785.0    47.2      0.89     2.22             15.8   44.4   \n",
       "4    [1, 0]   32.9  2279.0    66.7      0.89     2.44              6.7   66.7   \n",
       "5    [0, 1]   32.8  2064.0    43.7      0.22     2.11             16.1   66.7   \n",
       "6    [1, 0]   32.1  1903.0    48.9      0.89     2.44             12.8   66.7   \n",
       "..      ...    ...     ...     ...       ...      ...              ...    ...   \n",
       "131  [1, 0]   33.5  2013.0    54.6      0.72     2.44             10.7   46.2   \n",
       "132  [0, 1]   32.2  1831.0    46.3      0.85     2.22             15.2   48.8   \n",
       "133  [0, 1]   31.7  1797.0    47.6      0.84     2.47             11.7   57.9   \n",
       "134  [0, 1]   32.4  1803.0    52.4      1.05     2.41             11.4   64.1   \n",
       "135  [1, 0]   32.8  1867.0    50.6      0.98     2.46             12.1   43.9   \n",
       "\n",
       "     FT%_x  GD@15_x  ...  NASH%_y  NASHPG_y  Season_y  TD@15_y  \\\n",
       "1     44.4    620.0  ...     66.7      0.89      12.0      0.6   \n",
       "3     22.2  -2006.0  ...     66.7      0.89      12.0      0.6   \n",
       "4     77.8   1707.0  ...     66.7      0.89      12.0      0.6   \n",
       "5     66.7    397.0  ...     66.7      0.89      12.0      0.6   \n",
       "6     44.4    -60.0  ...     66.7      0.89      12.0      0.6   \n",
       "..     ...      ...  ...      ...       ...       ...      ...   \n",
       "131   41.0    245.0  ...     45.1      0.56      10.0      0.0   \n",
       "132   29.3   -619.0  ...     45.1      0.56      10.0      0.0   \n",
       "133   50.0    143.0  ...     45.1      0.56      10.0      0.0   \n",
       "134   48.7    927.0  ...     45.1      0.56      10.0      0.0   \n",
       "135   58.5    565.0  ...     45.1      0.56      10.0      0.0   \n",
       "\n",
       "     Towers killed_y  Towers lost_y  VWPM_y  WCPM_y  WPM_y  Win rate_y  \n",
       "1                6.9            5.6    1.20    1.44    3.4        66.7  \n",
       "3                6.9            5.6    1.20    1.44    3.4        66.7  \n",
       "4                6.9            5.6    1.20    1.44    3.4        66.7  \n",
       "5                6.9            5.6    1.20    1.44    3.4        66.7  \n",
       "6                6.9            5.6    1.20    1.44    3.4        66.7  \n",
       "..               ...            ...     ...     ...    ...         ...  \n",
       "131              6.1            6.0    1.48    1.51    3.4        53.7  \n",
       "132              6.1            6.0    1.48    1.51    3.4        53.7  \n",
       "133              6.1            6.0    1.48    1.51    3.4        53.7  \n",
       "134              6.1            6.0    1.48    1.51    3.4        53.7  \n",
       "135              6.1            6.0    1.48    1.51    3.4        53.7  \n",
       "\n",
       "[1477 rows x 55 columns]"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final[\"FT%_y\"] = final[\"FT%_y\"].replace('-', float(0)).astype(float)\n",
    "final[\"FT%_x\"] = final[\"FT%_x\"].replace('-', float(0)).astype(float)\n",
    "\n",
    "final[\"FB%_y\"] = final[\"FB%_y\"].replace('-', float(0)).astype(float)\n",
    "final[\"FB%_x\"] = final[\"FB%_x\"].replace('-', float(0)).astype(float)\n",
    "\n",
    "final[\"Season_y\"] = final[\"Season_y\"].apply(lambda x: str(x).replace('S', '')).astype(float)\n",
    "final[\"Season_x\"] = final[\"Season_x\"].apply(lambda x: str(x).replace('S', '')).astype(float)\n",
    "\n",
    "final = final.rename(columns={'Unnamed: 1': 'Team 1', 'Unnamed: 3': 'Team 2'})\n",
    "\n",
    "final[\"Score\"] = final[\"Score\"].apply(lambda x: formatScore(x))\n",
    "\n",
    "final[\"Win rate_x\"] = final[\"Win rate_x\"].apply(lambda x: float(x.replace(\"%\", \"\")))\n",
    "final[\"Win rate_y\"] = final[\"Win rate_y\"].apply(lambda x: float(x.replace(\"%\", \"\")))\n",
    "\n",
    "final[\"Game duration_x\"] = final[\"Game duration_x\"].apply(lambda x: get_sec(x))\n",
    "final[\"Game duration_y\"] = final[\"Game duration_y\"].apply(lambda x: get_sec(x))\n",
    "\n",
    "final = final.drop(columns=['Unnamed: 4', 'Patch', 'Date', 'Region_y', 'Region_x', 'Game', 'Team 1', 'Team 2', 'Name_y', 'Name_x'])\n",
    "final"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And this is what the final dataframe looks like.\n",
    "\n",
    "Next I will split the dataset into training, testing and validation sets. I also create one dimensional label array, but will leave the two dimensional in a variable in case I will utilize it for something later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set: 64.0 % of the original dataset\n",
      "Validation set: 16.0 % of the original dataset\n",
      "Test set: 20.0 % of the original dataset\n"
     ]
    }
   ],
   "source": [
    "np.array(final['Score'])\n",
    "resultArray = np.stack(final['Score'], axis=0)\n",
    "\n",
    "data = final.copy()\n",
    "data = data.drop(\"Score\", axis = 1)\n",
    "labels = resultArray\n",
    "onedLabels = final[\"Score\"].apply(lambda x: x[0])\n",
    "\n",
    "train_data, test_data, train_labels, test_labels = train_test_split(data, onedLabels, test_size=0.20, train_size=0.80, random_state=0, shuffle=True)\n",
    "train_data, validation_data, train_labels, validation_labels = train_test_split(train_data, train_labels, test_size=0.20, train_size=0.80, random_state=0)\n",
    "\n",
    "train_percentage = (train_data.count()[0] / data.count()[0]).round(2) * 100\n",
    "validation_percentage = (validation_data.count()[0] / data.count()[0]).round(2) * 100\n",
    "test_percentage = (test_data.count()[0] / data.count()[0]).round(2) * 100\n",
    "\n",
    "print (f\"Training set: {train_percentage} % of the original dataset\")\n",
    "print (f\"Validation set: {validation_percentage} % of the original dataset\")\n",
    "print (f\"Test set: {test_percentage} % of the original dataset\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Blue team win rate: 48.21 %\n",
      "Red team win rate: 51.79 %\n",
      "\n"
     ]
    }
   ],
   "source": [
    "blue, red = np.bincount(onedLabels)\n",
    "total = blue + red\n",
    "\n",
    "print(f\"Blue team win rate: {round(100 * blue / total, 2)} %\")\n",
    "print(f\"Red team win rate: {round(100 * red / total, 2)} %\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These prints are actually wrong and should be vice versa, so **Blue team: 51.79%** and **Red team 48.21%**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b381258-ac86-4c98-b563-e6f7fcc00ea4",
   "metadata": {
    "tags": []
   },
   "source": [
    "## 4. Modeling\n",
    "\n",
    "I decided to try bunch of different models and algorithms to analyze the data. Some of them fit well and some of them worse. In this section I'm analyzing different models for this particular machine learning problem and lastly will choose the final one to use in the demo section."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.1 GaussianNB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "71.62%\n"
     ]
    }
   ],
   "source": [
    "gaussianNB = GaussianNB()\n",
    "onedLabels = final[\"Score\"].apply(lambda x: x[0])\n",
    "\n",
    "gaussianNB.fit(train_data, train_labels)\n",
    "predictions = gaussianNB.predict(test_data)\n",
    "asd = gaussianNB.predict_proba(test_data)\n",
    "accuracy = accuracy_score(test_labels, predictions)\n",
    "print(f'{accuracy * 100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "GaussianNB seemed to fit well into this dataset, getting a respectable 75% accuracy. However, demoing the predictions as probabilities seemed to give confidently very unrealistic results. It seems that the model is overfitting, and thus I will not use it in the final demo."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.2 MultinomialNB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "61.15%\n"
     ]
    }
   ],
   "source": [
    "multinomialNB = MultinomialNB()\n",
    "train_dataNB = train_data.abs()\n",
    "test_dataNB = test_data.abs()\n",
    "train_labelsNB = train_labels.abs()\n",
    "test_labelsNB = test_labels.abs()\n",
    "\n",
    "multinomialNB.fit(train_dataNB, train_labelsNB)\n",
    "predictions = multinomialNB.predict(test_dataNB)\n",
    "accuracy = accuracy_score(test_labelsNB, predictions)\n",
    "print(f'{accuracy * 100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "MultinomialNB is used mainly in natural language classification and does not fit into this dataset. I still gave it a try just for curiosity."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.3 Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\r\n<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\r\n \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\r\n<!-- Generated by graphviz version 2.38.0 (20140413.2041)\r\n -->\r\n<!-- Title: Tree Pages: 1 -->\r\n<svg width=\"5458pt\" height=\"671pt\"\r\n viewBox=\"0.00 0.00 5458.00 671.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\r\n<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 667)\">\r\n<title>Tree</title>\r\n<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-667 5454,-667 5454,4 -4,4\"/>\r\n<!-- 0 -->\r\n<g id=\"node1\" class=\"node\"><title>0</title>\r\n<path fill=\"#f1f8fd\" stroke=\"black\" d=\"M2892.5,-663C2892.5,-663 2764.5,-663 2764.5,-663 2758.5,-663 2752.5,-657 2752.5,-651 2752.5,-651 2752.5,-592 2752.5,-592 2752.5,-586 2758.5,-580 2764.5,-580 2764.5,-580 2892.5,-580 2892.5,-580 2898.5,-580 2904.5,-586 2904.5,-592 2904.5,-592 2904.5,-651 2904.5,-651 2904.5,-657 2898.5,-663 2892.5,-663\"/>\r\n<text text-anchor=\"start\" x=\"2768.5\" y=\"-647.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_y ≤ 53.75</text>\r\n<text text-anchor=\"start\" x=\"2791\" y=\"-632.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.499</text>\r\n<text text-anchor=\"start\" x=\"2777\" y=\"-617.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 1477</text>\r\n<text text-anchor=\"start\" x=\"2771.5\" y=\"-602.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [712, 765]</text>\r\n<text text-anchor=\"start\" x=\"2760.5\" y=\"-587.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 1 -->\r\n<g id=\"node2\" class=\"node\"><title>1</title>\r\n<path fill=\"#9acdf2\" stroke=\"black\" d=\"M2143.5,-544C2143.5,-544 2015.5,-544 2015.5,-544 2009.5,-544 2003.5,-538 2003.5,-532 2003.5,-532 2003.5,-473 2003.5,-473 2003.5,-467 2009.5,-461 2015.5,-461 2015.5,-461 2143.5,-461 2143.5,-461 2149.5,-461 2155.5,-467 2155.5,-473 2155.5,-473 2155.5,-532 2155.5,-532 2155.5,-538 2149.5,-544 2143.5,-544\"/>\r\n<text text-anchor=\"start\" x=\"2019.5\" y=\"-528.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_x ≤ 44.55</text>\r\n<text text-anchor=\"start\" x=\"2042\" y=\"-513.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.441</text>\r\n<text text-anchor=\"start\" x=\"2032\" y=\"-498.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 835</text>\r\n<text text-anchor=\"start\" x=\"2022.5\" y=\"-483.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [274, 561]</text>\r\n<text text-anchor=\"start\" x=\"2011.5\" y=\"-468.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 0&#45;&gt;1 -->\r\n<g id=\"edge1\" class=\"edge\"><title>0&#45;&gt;1</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2752.21,-608.583C2611.7,-586.635 2312.93,-539.963 2165.48,-516.93\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2165.98,-513.466 2155.56,-515.38 2164.9,-520.382 2165.98,-513.466\"/>\r\n<text text-anchor=\"middle\" x=\"2170.29\" y=\"-531.875\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">True</text>\r\n</g>\r\n<!-- 32 -->\r\n<g id=\"node33\" class=\"node\"><title>32</title>\r\n<path fill=\"#f1bc95\" stroke=\"black\" d=\"M3475.5,-544C3475.5,-544 3341.5,-544 3341.5,-544 3335.5,-544 3329.5,-538 3329.5,-532 3329.5,-532 3329.5,-473 3329.5,-473 3329.5,-467 3335.5,-461 3341.5,-461 3341.5,-461 3475.5,-461 3475.5,-461 3481.5,-461 3487.5,-467 3487.5,-473 3487.5,-473 3487.5,-532 3487.5,-532 3487.5,-538 3481.5,-544 3475.5,-544\"/>\r\n<text text-anchor=\"start\" x=\"3337.5\" y=\"-528.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Towers killed_x ≤ 5.85</text>\r\n<text text-anchor=\"start\" x=\"3371\" y=\"-513.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.434</text>\r\n<text text-anchor=\"start\" x=\"3361\" y=\"-498.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 642</text>\r\n<text text-anchor=\"start\" x=\"3351.5\" y=\"-483.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [438, 204]</text>\r\n<text text-anchor=\"start\" x=\"3340\" y=\"-468.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 0&#45;&gt;32 -->\r\n<g id=\"edge32\" class=\"edge\"><title>0&#45;&gt;32</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2904.62,-605.144C3011.57,-583.569 3205.91,-544.366 3319.44,-521.465\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3320.39,-524.845 3329.5,-519.437 3319,-517.984 3320.39,-524.845\"/>\r\n<text text-anchor=\"middle\" x=\"3315.66\" y=\"-536.561\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">False</text>\r\n</g>\r\n<!-- 2 -->\r\n<g id=\"node3\" class=\"node\"><title>2</title>\r\n<path fill=\"#fcefe6\" stroke=\"black\" d=\"M1081,-425C1081,-425 952,-425 952,-425 946,-425 940,-419 940,-413 940,-413 940,-354 940,-354 940,-348 946,-342 952,-342 952,-342 1081,-342 1081,-342 1087,-342 1093,-348 1093,-354 1093,-354 1093,-413 1093,-413 1093,-419 1087,-425 1081,-425\"/>\r\n<text text-anchor=\"start\" x=\"960.5\" y=\"-409.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_y ≤ 38.5</text>\r\n<text text-anchor=\"start\" x=\"979\" y=\"-394.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.498</text>\r\n<text text-anchor=\"start\" x=\"969\" y=\"-379.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 330</text>\r\n<text text-anchor=\"start\" x=\"959.5\" y=\"-364.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [176, 154]</text>\r\n<text text-anchor=\"start\" x=\"948\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 1&#45;&gt;2 -->\r\n<g id=\"edge2\" class=\"edge\"><title>1&#45;&gt;2</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2003.48,-493.133C1811.19,-471.968 1306.31,-416.398 1103.08,-394.03\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1103.38,-390.541 1093.05,-392.926 1102.61,-397.499 1103.38,-390.541\"/>\r\n</g>\r\n<!-- 17 -->\r\n<g id=\"node18\" class=\"node\"><title>17</title>\r\n<path fill=\"#69b5eb\" stroke=\"black\" d=\"M2143.5,-425C2143.5,-425 2015.5,-425 2015.5,-425 2009.5,-425 2003.5,-419 2003.5,-413 2003.5,-413 2003.5,-354 2003.5,-354 2003.5,-348 2009.5,-342 2015.5,-342 2015.5,-342 2143.5,-342 2143.5,-342 2149.5,-342 2155.5,-348 2155.5,-354 2155.5,-354 2155.5,-413 2155.5,-413 2155.5,-419 2149.5,-425 2143.5,-425\"/>\r\n<text text-anchor=\"start\" x=\"2038\" y=\"-409.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">K:D_y ≤ 0.97</text>\r\n<text text-anchor=\"start\" x=\"2042\" y=\"-394.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.313</text>\r\n<text text-anchor=\"start\" x=\"2032\" y=\"-379.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 505</text>\r\n<text text-anchor=\"start\" x=\"2026.5\" y=\"-364.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [98, 407]</text>\r\n<text text-anchor=\"start\" x=\"2011.5\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 1&#45;&gt;17 -->\r\n<g id=\"edge17\" class=\"edge\"><title>1&#45;&gt;17</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2079.5,-460.907C2079.5,-452.649 2079.5,-443.864 2079.5,-435.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2083,-435.021 2079.5,-425.021 2076,-435.021 2083,-435.021\"/>\r\n</g>\r\n<!-- 3 -->\r\n<g id=\"node4\" class=\"node\"><title>3</title>\r\n<path fill=\"#a7d3f3\" stroke=\"black\" d=\"M567.5,-306C567.5,-306 439.5,-306 439.5,-306 433.5,-306 427.5,-300 427.5,-294 427.5,-294 427.5,-235 427.5,-235 427.5,-229 433.5,-223 439.5,-223 439.5,-223 567.5,-223 567.5,-223 573.5,-223 579.5,-229 579.5,-235 579.5,-235 579.5,-294 579.5,-294 579.5,-300 573.5,-306 567.5,-306\"/>\r\n<text text-anchor=\"start\" x=\"447.5\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_x ≤ 28.7</text>\r\n<text text-anchor=\"start\" x=\"466\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.459</text>\r\n<text text-anchor=\"start\" x=\"456\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 140</text>\r\n<text text-anchor=\"start\" x=\"455\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [50, 90]</text>\r\n<text text-anchor=\"start\" x=\"435.5\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 2&#45;&gt;3 -->\r\n<g id=\"edge3\" class=\"edge\"><title>2&#45;&gt;3</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M939.976,-365.047C846.137,-343.645 687.516,-307.469 589.533,-285.122\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"590.186,-281.681 579.658,-282.869 588.629,-288.505 590.186,-281.681\"/>\r\n</g>\r\n<!-- 10 -->\r\n<g id=\"node11\" class=\"node\"><title>10</title>\r\n<path fill=\"#f2c19e\" stroke=\"black\" d=\"M1097,-306C1097,-306 936,-306 936,-306 930,-306 924,-300 924,-294 924,-294 924,-235 924,-235 924,-229 930,-223 936,-223 936,-223 1097,-223 1097,-223 1103,-223 1109,-229 1109,-235 1109,-235 1109,-294 1109,-294 1109,-300 1103,-306 1097,-306\"/>\r\n<text text-anchor=\"start\" x=\"932\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Game duration_y ≤ 2014.5</text>\r\n<text text-anchor=\"start\" x=\"979\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.447</text>\r\n<text text-anchor=\"start\" x=\"969\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 190</text>\r\n<text text-anchor=\"start\" x=\"963.5\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [126, 64]</text>\r\n<text text-anchor=\"start\" x=\"948\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 2&#45;&gt;10 -->\r\n<g id=\"edge10\" class=\"edge\"><title>2&#45;&gt;10</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1016.5,-341.907C1016.5,-333.649 1016.5,-324.864 1016.5,-316.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1020,-316.021 1016.5,-306.021 1013,-316.021 1020,-316.021\"/>\r\n</g>\r\n<!-- 4 -->\r\n<g id=\"node5\" class=\"node\"><title>4</title>\r\n<path fill=\"#eb9c64\" stroke=\"black\" d=\"M312,-187C312,-187 183,-187 183,-187 177,-187 171,-181 171,-175 171,-175 171,-116 171,-116 171,-110 177,-104 183,-104 183,-104 312,-104 312,-104 318,-104 324,-110 324,-116 324,-116 324,-175 324,-175 324,-181 318,-187 312,-187\"/>\r\n<text text-anchor=\"start\" x=\"202\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">K:D_y ≤ 0.695</text>\r\n<text text-anchor=\"start\" x=\"210\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.293</text>\r\n<text text-anchor=\"start\" x=\"204\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 28</text>\r\n<text text-anchor=\"start\" x=\"203\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [23, 5]</text>\r\n<text text-anchor=\"start\" x=\"179\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 3&#45;&gt;4 -->\r\n<g id=\"edge4\" class=\"edge\"><title>3&#45;&gt;4</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M427.385,-228.713C397.866,-215.222 363.916,-199.706 333.518,-185.813\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"334.757,-182.531 324.207,-181.557 331.847,-188.898 334.757,-182.531\"/>\r\n</g>\r\n<!-- 7 -->\r\n<g id=\"node8\" class=\"node\"><title>7</title>\r\n<path fill=\"#78bced\" stroke=\"black\" d=\"M567.5,-187C567.5,-187 439.5,-187 439.5,-187 433.5,-187 427.5,-181 427.5,-175 427.5,-175 427.5,-116 427.5,-116 427.5,-110 433.5,-104 439.5,-104 439.5,-104 567.5,-104 567.5,-104 573.5,-104 579.5,-110 579.5,-116 579.5,-116 579.5,-175 579.5,-175 579.5,-181 573.5,-187 567.5,-187\"/>\r\n<text text-anchor=\"start\" x=\"456.5\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">WPM_x ≤ 3.25</text>\r\n<text text-anchor=\"start\" x=\"466\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.366</text>\r\n<text text-anchor=\"start\" x=\"456\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 112</text>\r\n<text text-anchor=\"start\" x=\"455\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [27, 85]</text>\r\n<text text-anchor=\"start\" x=\"435.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 3&#45;&gt;7 -->\r\n<g id=\"edge7\" class=\"edge\"><title>3&#45;&gt;7</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M503.5,-222.907C503.5,-214.649 503.5,-205.864 503.5,-197.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"507,-197.021 503.5,-187.021 500,-197.021 507,-197.021\"/>\r\n</g>\r\n<!-- 5 -->\r\n<g id=\"node6\" class=\"node\"><title>5</title>\r\n<path fill=\"#f5d0b5\" stroke=\"black\" d=\"M141,-68C141,-68 12,-68 12,-68 6,-68 0,-62 0,-56 0,-56 0,-12 0,-12 0,-6 6,-0 12,-0 12,-0 141,-0 141,-0 147,-0 153,-6 153,-12 153,-12 153,-56 153,-56 153,-62 147,-68 141,-68\"/>\r\n<text text-anchor=\"start\" x=\"39\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.473</text>\r\n<text text-anchor=\"start\" x=\"33\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 13</text>\r\n<text text-anchor=\"start\" x=\"36\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [8, 5]</text>\r\n<text text-anchor=\"start\" x=\"8\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 4&#45;&gt;5 -->\r\n<g id=\"edge5\" class=\"edge\"><title>4&#45;&gt;5</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M183.826,-103.726C168.506,-93.9161 152.184,-83.4644 137.013,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"138.539,-70.5704 128.23,-68.1252 134.764,-76.4654 138.539,-70.5704\"/>\r\n</g>\r\n<!-- 6 -->\r\n<g id=\"node7\" class=\"node\"><title>6</title>\r\n<path fill=\"#e58139\" stroke=\"black\" d=\"M312,-68C312,-68 183,-68 183,-68 177,-68 171,-62 171,-56 171,-56 171,-12 171,-12 171,-6 177,-0 183,-0 183,-0 312,-0 312,-0 318,-0 324,-6 324,-12 324,-12 324,-56 324,-56 324,-62 318,-68 312,-68\"/>\r\n<text text-anchor=\"start\" x=\"218.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"204\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 15</text>\r\n<text text-anchor=\"start\" x=\"203\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [15, 0]</text>\r\n<text text-anchor=\"start\" x=\"179\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 4&#45;&gt;6 -->\r\n<g id=\"edge6\" class=\"edge\"><title>4&#45;&gt;6</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M247.5,-103.726C247.5,-95.5175 247.5,-86.8595 247.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"251,-78.2996 247.5,-68.2996 244,-78.2996 251,-78.2996\"/>\r\n</g>\r\n<!-- 8 -->\r\n<g id=\"node9\" class=\"node\"><title>8</title>\r\n<path fill=\"#d9ecfa\" stroke=\"black\" d=\"M482.5,-68C482.5,-68 354.5,-68 354.5,-68 348.5,-68 342.5,-62 342.5,-56 342.5,-56 342.5,-12 342.5,-12 342.5,-6 348.5,-0 354.5,-0 354.5,-0 482.5,-0 482.5,-0 488.5,-0 494.5,-6 494.5,-12 494.5,-12 494.5,-56 494.5,-56 494.5,-62 488.5,-68 482.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"381\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.494</text>\r\n<text text-anchor=\"start\" x=\"375\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 38</text>\r\n<text text-anchor=\"start\" x=\"370\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [17, 21]</text>\r\n<text text-anchor=\"start\" x=\"350.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 7&#45;&gt;8 -->\r\n<g id=\"edge8\" class=\"edge\"><title>7&#45;&gt;8</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M471.849,-103.726C464.911,-94.7878 457.559,-85.3168 450.603,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"453.246,-74.0528 444.349,-68.2996 447.716,-78.3451 453.246,-74.0528\"/>\r\n</g>\r\n<!-- 9 -->\r\n<g id=\"node10\" class=\"node\"><title>9</title>\r\n<path fill=\"#58ace9\" stroke=\"black\" d=\"M652.5,-68C652.5,-68 524.5,-68 524.5,-68 518.5,-68 512.5,-62 512.5,-56 512.5,-56 512.5,-12 512.5,-12 512.5,-6 518.5,-0 524.5,-0 524.5,-0 652.5,-0 652.5,-0 658.5,-0 664.5,-6 664.5,-12 664.5,-12 664.5,-56 664.5,-56 664.5,-62 658.5,-68 652.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"551\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.234</text>\r\n<text text-anchor=\"start\" x=\"545\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 74</text>\r\n<text text-anchor=\"start\" x=\"540\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [10, 64]</text>\r\n<text text-anchor=\"start\" x=\"520.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 7&#45;&gt;9 -->\r\n<g id=\"edge9\" class=\"edge\"><title>7&#45;&gt;9</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M535.151,-103.726C542.089,-94.7878 549.441,-85.3168 556.397,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"559.284,-78.3451 562.651,-68.2996 553.754,-74.0528 559.284,-78.3451\"/>\r\n</g>\r\n<!-- 11 -->\r\n<g id=\"node12\" class=\"node\"><title>11</title>\r\n<path fill=\"#edaa7a\" stroke=\"black\" d=\"M995,-187C995,-187 866,-187 866,-187 860,-187 854,-181 854,-175 854,-175 854,-116 854,-116 854,-110 860,-104 866,-104 866,-104 995,-104 995,-104 1001,-104 1007,-110 1007,-116 1007,-116 1007,-175 1007,-175 1007,-181 1001,-187 995,-187\"/>\r\n<text text-anchor=\"start\" x=\"881\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">CSM_x ≤ 32.45</text>\r\n<text text-anchor=\"start\" x=\"893\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.371</text>\r\n<text text-anchor=\"start\" x=\"883\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 122</text>\r\n<text text-anchor=\"start\" x=\"882\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [92, 30]</text>\r\n<text text-anchor=\"start\" x=\"862\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 10&#45;&gt;11 -->\r\n<g id=\"edge11\" class=\"edge\"><title>10&#45;&gt;11</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M986.663,-222.907C980.126,-214.014 973.139,-204.509 966.393,-195.331\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"969.027,-193.005 960.285,-187.021 963.387,-197.151 969.027,-193.005\"/>\r\n</g>\r\n<!-- 14 -->\r\n<g id=\"node15\" class=\"node\"><title>14</title>\r\n<path fill=\"#ffffff\" stroke=\"black\" d=\"M1166,-187C1166,-187 1037,-187 1037,-187 1031,-187 1025,-181 1025,-175 1025,-175 1025,-116 1025,-116 1025,-110 1031,-104 1037,-104 1037,-104 1166,-104 1166,-104 1172,-104 1178,-110 1178,-116 1178,-116 1178,-175 1178,-175 1178,-181 1172,-187 1166,-187\"/>\r\n<text text-anchor=\"start\" x=\"1052\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">FB%_x ≤ 47.75</text>\r\n<text text-anchor=\"start\" x=\"1072.5\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.5</text>\r\n<text text-anchor=\"start\" x=\"1058\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 68</text>\r\n<text text-anchor=\"start\" x=\"1053\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [34, 34]</text>\r\n<text text-anchor=\"start\" x=\"1033\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 10&#45;&gt;14 -->\r\n<g id=\"edge14\" class=\"edge\"><title>10&#45;&gt;14</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1045.99,-222.907C1052.45,-214.014 1059.36,-204.509 1066.02,-195.331\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1069.02,-197.168 1072.06,-187.021 1063.35,-193.054 1069.02,-197.168\"/>\r\n</g>\r\n<!-- 12 -->\r\n<g id=\"node13\" class=\"node\"><title>12</title>\r\n<path fill=\"#ea995f\" stroke=\"black\" d=\"M824,-68C824,-68 695,-68 695,-68 689,-68 683,-62 683,-56 683,-56 683,-12 683,-12 683,-6 689,-0 695,-0 695,-0 824,-0 824,-0 830,-0 836,-6 836,-12 836,-12 836,-56 836,-56 836,-62 830,-68 824,-68\"/>\r\n<text text-anchor=\"start\" x=\"722\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.269</text>\r\n<text text-anchor=\"start\" x=\"716\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 75</text>\r\n<text text-anchor=\"start\" x=\"711\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [63, 12]</text>\r\n<text text-anchor=\"start\" x=\"691\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 11&#45;&gt;12 -->\r\n<g id=\"edge12\" class=\"edge\"><title>11&#45;&gt;12</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M866.826,-103.726C851.506,-93.9161 835.184,-83.4644 820.013,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"821.539,-70.5704 811.23,-68.1252 817.764,-76.4654 821.539,-70.5704\"/>\r\n</g>\r\n<!-- 13 -->\r\n<g id=\"node14\" class=\"node\"><title>13</title>\r\n<path fill=\"#f5cfb4\" stroke=\"black\" d=\"M995,-68C995,-68 866,-68 866,-68 860,-68 854,-62 854,-56 854,-56 854,-12 854,-12 854,-6 860,-0 866,-0 866,-0 995,-0 995,-0 1001,-0 1007,-6 1007,-12 1007,-12 1007,-56 1007,-56 1007,-62 1001,-68 995,-68\"/>\r\n<text text-anchor=\"start\" x=\"893\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.473</text>\r\n<text text-anchor=\"start\" x=\"887\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 47</text>\r\n<text text-anchor=\"start\" x=\"882\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [29, 18]</text>\r\n<text text-anchor=\"start\" x=\"862\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 11&#45;&gt;13 -->\r\n<g id=\"edge13\" class=\"edge\"><title>11&#45;&gt;13</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M930.5,-103.726C930.5,-95.5175 930.5,-86.8595 930.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"934,-78.2996 930.5,-68.2996 927,-78.2996 934,-78.2996\"/>\r\n</g>\r\n<!-- 15 -->\r\n<g id=\"node16\" class=\"node\"><title>15</title>\r\n<path fill=\"#a9d4f4\" stroke=\"black\" d=\"M1165.5,-68C1165.5,-68 1037.5,-68 1037.5,-68 1031.5,-68 1025.5,-62 1025.5,-56 1025.5,-56 1025.5,-12 1025.5,-12 1025.5,-6 1031.5,-0 1037.5,-0 1037.5,-0 1165.5,-0 1165.5,-0 1171.5,-0 1177.5,-6 1177.5,-12 1177.5,-12 1177.5,-56 1177.5,-56 1177.5,-62 1171.5,-68 1165.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"1064\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.461</text>\r\n<text text-anchor=\"start\" x=\"1058\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 36</text>\r\n<text text-anchor=\"start\" x=\"1053\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [13, 23]</text>\r\n<text text-anchor=\"start\" x=\"1033.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 14&#45;&gt;15 -->\r\n<g id=\"edge15\" class=\"edge\"><title>14&#45;&gt;15</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1101.5,-103.726C1101.5,-95.5175 1101.5,-86.8595 1101.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1105,-78.2996 1101.5,-68.2996 1098,-78.2996 1105,-78.2996\"/>\r\n</g>\r\n<!-- 16 -->\r\n<g id=\"node17\" class=\"node\"><title>16</title>\r\n<path fill=\"#f3c3a1\" stroke=\"black\" d=\"M1337,-68C1337,-68 1208,-68 1208,-68 1202,-68 1196,-62 1196,-56 1196,-56 1196,-12 1196,-12 1196,-6 1202,-0 1208,-0 1208,-0 1337,-0 1337,-0 1343,-0 1349,-6 1349,-12 1349,-12 1349,-56 1349,-56 1349,-62 1343,-68 1337,-68\"/>\r\n<text text-anchor=\"start\" x=\"1235\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.451</text>\r\n<text text-anchor=\"start\" x=\"1229\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 32</text>\r\n<text text-anchor=\"start\" x=\"1224\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [21, 11]</text>\r\n<text text-anchor=\"start\" x=\"1204\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 14&#45;&gt;16 -->\r\n<g id=\"edge16\" class=\"edge\"><title>14&#45;&gt;16</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1165.17,-103.726C1180.49,-93.9161 1196.82,-83.4644 1211.99,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1214.24,-76.4654 1220.77,-68.1252 1210.46,-70.5704 1214.24,-76.4654\"/>\r\n</g>\r\n<!-- 18 -->\r\n<g id=\"node19\" class=\"node\"><title>18</title>\r\n<path fill=\"#5caeea\" stroke=\"black\" d=\"M1933.5,-306C1933.5,-306 1805.5,-306 1805.5,-306 1799.5,-306 1793.5,-300 1793.5,-294 1793.5,-294 1793.5,-235 1793.5,-235 1793.5,-229 1799.5,-223 1805.5,-223 1805.5,-223 1933.5,-223 1933.5,-223 1939.5,-223 1945.5,-229 1945.5,-235 1945.5,-235 1945.5,-294 1945.5,-294 1945.5,-300 1939.5,-306 1933.5,-306\"/>\r\n<text text-anchor=\"start\" x=\"1824\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">K:D_x ≤ 1.415</text>\r\n<text text-anchor=\"start\" x=\"1832\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.256</text>\r\n<text text-anchor=\"start\" x=\"1822\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 391</text>\r\n<text text-anchor=\"start\" x=\"1816.5\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [59, 332]</text>\r\n<text text-anchor=\"start\" x=\"1801.5\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 17&#45;&gt;18 -->\r\n<g id=\"edge18\" class=\"edge\"><title>17&#45;&gt;18</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2006.64,-341.907C1988.72,-331.925 1969.42,-321.172 1951.12,-310.976\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1952.67,-307.83 1942.23,-306.021 1949.26,-313.945 1952.67,-307.83\"/>\r\n</g>\r\n<!-- 25 -->\r\n<g id=\"node26\" class=\"node\"><title>25</title>\r\n<path fill=\"#a0d0f3\" stroke=\"black\" d=\"M2268.5,-306C2268.5,-306 2140.5,-306 2140.5,-306 2134.5,-306 2128.5,-300 2128.5,-294 2128.5,-294 2128.5,-235 2128.5,-235 2128.5,-229 2134.5,-223 2140.5,-223 2140.5,-223 2268.5,-223 2268.5,-223 2274.5,-223 2280.5,-229 2280.5,-235 2280.5,-235 2280.5,-294 2280.5,-294 2280.5,-300 2274.5,-306 2268.5,-306\"/>\r\n<text text-anchor=\"start\" x=\"2144.5\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_x ≤ 67.05</text>\r\n<text text-anchor=\"start\" x=\"2171\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.45</text>\r\n<text text-anchor=\"start\" x=\"2157\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 114</text>\r\n<text text-anchor=\"start\" x=\"2156\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [39, 75]</text>\r\n<text text-anchor=\"start\" x=\"2136.5\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 17&#45;&gt;25 -->\r\n<g id=\"edge25\" class=\"edge\"><title>17&#45;&gt;25</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2122.87,-341.907C2132.85,-332.56 2143.56,-322.538 2153.83,-312.929\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2156.3,-315.41 2161.21,-306.021 2151.52,-310.299 2156.3,-315.41\"/>\r\n</g>\r\n<!-- 19 -->\r\n<g id=\"node20\" class=\"node\"><title>19</title>\r\n<path fill=\"#67b4eb\" stroke=\"black\" d=\"M1678.5,-187C1678.5,-187 1550.5,-187 1550.5,-187 1544.5,-187 1538.5,-181 1538.5,-175 1538.5,-175 1538.5,-116 1538.5,-116 1538.5,-110 1544.5,-104 1550.5,-104 1550.5,-104 1678.5,-104 1678.5,-104 1684.5,-104 1690.5,-110 1690.5,-116 1690.5,-116 1690.5,-175 1690.5,-175 1690.5,-181 1684.5,-187 1678.5,-187\"/>\r\n<text text-anchor=\"start\" x=\"1565.5\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">FT%_y ≤ 11.95</text>\r\n<text text-anchor=\"start\" x=\"1577\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.307</text>\r\n<text text-anchor=\"start\" x=\"1567\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 301</text>\r\n<text text-anchor=\"start\" x=\"1561.5\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [57, 244]</text>\r\n<text text-anchor=\"start\" x=\"1546.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 18&#45;&gt;19 -->\r\n<g id=\"edge19\" class=\"edge\"><title>18&#45;&gt;19</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1793.34,-228.555C1763.94,-215.069 1730.18,-199.577 1699.97,-185.714\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1701.26,-182.456 1690.71,-181.467 1698.34,-188.819 1701.26,-182.456\"/>\r\n</g>\r\n<!-- 22 -->\r\n<g id=\"node23\" class=\"node\"><title>22</title>\r\n<path fill=\"#3d9fe6\" stroke=\"black\" d=\"M1933.5,-187C1933.5,-187 1805.5,-187 1805.5,-187 1799.5,-187 1793.5,-181 1793.5,-175 1793.5,-175 1793.5,-116 1793.5,-116 1793.5,-110 1799.5,-104 1805.5,-104 1805.5,-104 1933.5,-104 1933.5,-104 1939.5,-104 1945.5,-110 1945.5,-116 1945.5,-116 1945.5,-175 1945.5,-175 1945.5,-181 1939.5,-187 1933.5,-187\"/>\r\n<text text-anchor=\"start\" x=\"1824\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">FB%_y ≤ 77.8</text>\r\n<text text-anchor=\"start\" x=\"1832\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.043</text>\r\n<text text-anchor=\"start\" x=\"1826\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 90</text>\r\n<text text-anchor=\"start\" x=\"1825\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [2, 88]</text>\r\n<text text-anchor=\"start\" x=\"1801.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 18&#45;&gt;22 -->\r\n<g id=\"edge22\" class=\"edge\"><title>18&#45;&gt;22</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1869.5,-222.907C1869.5,-214.649 1869.5,-205.864 1869.5,-197.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1873,-197.021 1869.5,-187.021 1866,-197.021 1873,-197.021\"/>\r\n</g>\r\n<!-- 20 -->\r\n<g id=\"node21\" class=\"node\"><title>20</title>\r\n<path fill=\"#eca06a\" stroke=\"black\" d=\"M1508,-68C1508,-68 1379,-68 1379,-68 1373,-68 1367,-62 1367,-56 1367,-56 1367,-12 1367,-12 1367,-6 1373,-0 1379,-0 1379,-0 1508,-0 1508,-0 1514,-0 1520,-6 1520,-12 1520,-12 1520,-56 1520,-56 1520,-62 1514,-68 1508,-68\"/>\r\n<text text-anchor=\"start\" x=\"1410\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.32</text>\r\n<text text-anchor=\"start\" x=\"1404\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 5</text>\r\n<text text-anchor=\"start\" x=\"1403\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [4, 1]</text>\r\n<text text-anchor=\"start\" x=\"1375\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 19&#45;&gt;20 -->\r\n<g id=\"edge20\" class=\"edge\"><title>19&#45;&gt;20</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1550.83,-103.726C1535.51,-93.9161 1519.18,-83.4644 1504.01,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1505.54,-70.5704 1495.23,-68.1252 1501.76,-76.4654 1505.54,-70.5704\"/>\r\n</g>\r\n<!-- 21 -->\r\n<g id=\"node22\" class=\"node\"><title>21</title>\r\n<path fill=\"#64b2eb\" stroke=\"black\" d=\"M1678.5,-68C1678.5,-68 1550.5,-68 1550.5,-68 1544.5,-68 1538.5,-62 1538.5,-56 1538.5,-56 1538.5,-12 1538.5,-12 1538.5,-6 1544.5,-0 1550.5,-0 1550.5,-0 1678.5,-0 1678.5,-0 1684.5,-0 1690.5,-6 1690.5,-12 1690.5,-12 1690.5,-56 1690.5,-56 1690.5,-62 1684.5,-68 1678.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"1577\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.294</text>\r\n<text text-anchor=\"start\" x=\"1567\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 296</text>\r\n<text text-anchor=\"start\" x=\"1561.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [53, 243]</text>\r\n<text text-anchor=\"start\" x=\"1546.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 19&#45;&gt;21 -->\r\n<g id=\"edge21\" class=\"edge\"><title>19&#45;&gt;21</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1614.5,-103.726C1614.5,-95.5175 1614.5,-86.8595 1614.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1618,-78.2996 1614.5,-68.2996 1611,-78.2996 1618,-78.2996\"/>\r\n</g>\r\n<!-- 23 -->\r\n<g id=\"node24\" class=\"node\"><title>23</title>\r\n<path fill=\"#3b9ee5\" stroke=\"black\" d=\"M1848.5,-68C1848.5,-68 1720.5,-68 1720.5,-68 1714.5,-68 1708.5,-62 1708.5,-56 1708.5,-56 1708.5,-12 1708.5,-12 1708.5,-6 1714.5,-0 1720.5,-0 1720.5,-0 1848.5,-0 1848.5,-0 1854.5,-0 1860.5,-6 1860.5,-12 1860.5,-12 1860.5,-56 1860.5,-56 1860.5,-62 1854.5,-68 1848.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"1747\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.022</text>\r\n<text text-anchor=\"start\" x=\"1741\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 89</text>\r\n<text text-anchor=\"start\" x=\"1740\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [1, 88]</text>\r\n<text text-anchor=\"start\" x=\"1716.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 22&#45;&gt;23 -->\r\n<g id=\"edge23\" class=\"edge\"><title>22&#45;&gt;23</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1837.85,-103.726C1830.91,-94.7878 1823.56,-85.3168 1816.6,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1819.25,-74.0528 1810.35,-68.2996 1813.72,-78.3451 1819.25,-74.0528\"/>\r\n</g>\r\n<!-- 24 -->\r\n<g id=\"node25\" class=\"node\"><title>24</title>\r\n<path fill=\"#e58139\" stroke=\"black\" d=\"M2020,-68C2020,-68 1891,-68 1891,-68 1885,-68 1879,-62 1879,-56 1879,-56 1879,-12 1879,-12 1879,-6 1885,-0 1891,-0 1891,-0 2020,-0 2020,-0 2026,-0 2032,-6 2032,-12 2032,-12 2032,-56 2032,-56 2032,-62 2026,-68 2020,-68\"/>\r\n<text text-anchor=\"start\" x=\"1926.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"1916\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 1</text>\r\n<text text-anchor=\"start\" x=\"1915\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [1, 0]</text>\r\n<text text-anchor=\"start\" x=\"1887\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 22&#45;&gt;24 -->\r\n<g id=\"edge24\" class=\"edge\"><title>22&#45;&gt;24</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M1901.52,-103.726C1908.54,-94.7878 1915.98,-85.3168 1923.02,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"1925.92,-78.3259 1929.35,-68.2996 1920.42,-74.0022 1925.92,-78.3259\"/>\r\n</g>\r\n<!-- 26 -->\r\n<g id=\"node27\" class=\"node\"><title>26</title>\r\n<path fill=\"#daedfa\" stroke=\"black\" d=\"M2285,-187C2285,-187 2124,-187 2124,-187 2118,-187 2112,-181 2112,-175 2112,-175 2112,-116 2112,-116 2112,-110 2118,-104 2124,-104 2124,-104 2285,-104 2285,-104 2291,-104 2297,-110 2297,-116 2297,-116 2297,-175 2297,-175 2297,-181 2291,-187 2285,-187\"/>\r\n<text text-anchor=\"start\" x=\"2120\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Game duration_x ≤ 2045.5</text>\r\n<text text-anchor=\"start\" x=\"2167\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.495</text>\r\n<text text-anchor=\"start\" x=\"2161\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 78</text>\r\n<text text-anchor=\"start\" x=\"2156\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [35, 43]</text>\r\n<text text-anchor=\"start\" x=\"2136.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 25&#45;&gt;26 -->\r\n<g id=\"edge26\" class=\"edge\"><title>25&#45;&gt;26</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2204.5,-222.907C2204.5,-214.649 2204.5,-205.864 2204.5,-197.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2208,-197.021 2204.5,-187.021 2201,-197.021 2208,-197.021\"/>\r\n</g>\r\n<!-- 29 -->\r\n<g id=\"node30\" class=\"node\"><title>29</title>\r\n<path fill=\"#52a9e8\" stroke=\"black\" d=\"M2532.5,-187C2532.5,-187 2404.5,-187 2404.5,-187 2398.5,-187 2392.5,-181 2392.5,-175 2392.5,-175 2392.5,-116 2392.5,-116 2392.5,-110 2398.5,-104 2404.5,-104 2404.5,-104 2532.5,-104 2532.5,-104 2538.5,-104 2544.5,-110 2544.5,-116 2544.5,-116 2544.5,-175 2544.5,-175 2544.5,-181 2538.5,-187 2532.5,-187\"/>\r\n<text text-anchor=\"start\" x=\"2416.5\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VWPM_x ≤ 1.02</text>\r\n<text text-anchor=\"start\" x=\"2431\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.198</text>\r\n<text text-anchor=\"start\" x=\"2425\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 36</text>\r\n<text text-anchor=\"start\" x=\"2424\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [4, 32]</text>\r\n<text text-anchor=\"start\" x=\"2400.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 25&#45;&gt;29 -->\r\n<g id=\"edge29\" class=\"edge\"><title>25&#45;&gt;29</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2280.51,-229.816C2312.53,-215.623 2349.95,-199.039 2382.94,-184.419\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2384.39,-187.603 2392.12,-180.352 2381.56,-181.204 2384.39,-187.603\"/>\r\n</g>\r\n<!-- 27 -->\r\n<g id=\"node28\" class=\"node\"><title>27</title>\r\n<path fill=\"#fcefe6\" stroke=\"black\" d=\"M2191,-68C2191,-68 2062,-68 2062,-68 2056,-68 2050,-62 2050,-56 2050,-56 2050,-12 2050,-12 2050,-6 2056,-0 2062,-0 2062,-0 2191,-0 2191,-0 2197,-0 2203,-6 2203,-12 2203,-12 2203,-56 2203,-56 2203,-62 2197,-68 2191,-68\"/>\r\n<text text-anchor=\"start\" x=\"2089\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.498</text>\r\n<text text-anchor=\"start\" x=\"2083\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 60</text>\r\n<text text-anchor=\"start\" x=\"2078\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [32, 28]</text>\r\n<text text-anchor=\"start\" x=\"2058\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 26&#45;&gt;27 -->\r\n<g id=\"edge27\" class=\"edge\"><title>26&#45;&gt;27</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2175.46,-103.726C2169.15,-94.879 2162.48,-85.51 2156.15,-76.6303\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2158.87,-74.4138 2150.22,-68.2996 2153.17,-78.4751 2158.87,-74.4138\"/>\r\n</g>\r\n<!-- 28 -->\r\n<g id=\"node29\" class=\"node\"><title>28</title>\r\n<path fill=\"#61b1ea\" stroke=\"black\" d=\"M2361.5,-68C2361.5,-68 2233.5,-68 2233.5,-68 2227.5,-68 2221.5,-62 2221.5,-56 2221.5,-56 2221.5,-12 2221.5,-12 2221.5,-6 2227.5,-0 2233.5,-0 2233.5,-0 2361.5,-0 2361.5,-0 2367.5,-0 2373.5,-6 2373.5,-12 2373.5,-12 2373.5,-56 2373.5,-56 2373.5,-62 2367.5,-68 2361.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"2260\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.278</text>\r\n<text text-anchor=\"start\" x=\"2254\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 18</text>\r\n<text text-anchor=\"start\" x=\"2253\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [3, 15]</text>\r\n<text text-anchor=\"start\" x=\"2229.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 26&#45;&gt;28 -->\r\n<g id=\"edge28\" class=\"edge\"><title>26&#45;&gt;28</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2239.13,-103.726C2246.8,-94.6966 2254.93,-85.1235 2262.61,-76.0816\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2265.41,-78.1873 2269.22,-68.2996 2260.08,-73.6559 2265.41,-78.1873\"/>\r\n</g>\r\n<!-- 30 -->\r\n<g id=\"node31\" class=\"node\"><title>30</title>\r\n<path fill=\"#e58139\" stroke=\"black\" d=\"M2533,-68C2533,-68 2404,-68 2404,-68 2398,-68 2392,-62 2392,-56 2392,-56 2392,-12 2392,-12 2392,-6 2398,-0 2404,-0 2404,-0 2533,-0 2533,-0 2539,-0 2545,-6 2545,-12 2545,-12 2545,-56 2545,-56 2545,-62 2539,-68 2533,-68\"/>\r\n<text text-anchor=\"start\" x=\"2439.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"2429\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 1</text>\r\n<text text-anchor=\"start\" x=\"2428\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [1, 0]</text>\r\n<text text-anchor=\"start\" x=\"2400\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 29&#45;&gt;30 -->\r\n<g id=\"edge30\" class=\"edge\"><title>29&#45;&gt;30</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2468.5,-103.726C2468.5,-95.5175 2468.5,-86.8595 2468.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2472,-78.2996 2468.5,-68.2996 2465,-78.2996 2472,-78.2996\"/>\r\n</g>\r\n<!-- 31 -->\r\n<g id=\"node32\" class=\"node\"><title>31</title>\r\n<path fill=\"#4ca6e7\" stroke=\"black\" d=\"M2703.5,-68C2703.5,-68 2575.5,-68 2575.5,-68 2569.5,-68 2563.5,-62 2563.5,-56 2563.5,-56 2563.5,-12 2563.5,-12 2563.5,-6 2569.5,-0 2575.5,-0 2575.5,-0 2703.5,-0 2703.5,-0 2709.5,-0 2715.5,-6 2715.5,-12 2715.5,-12 2715.5,-56 2715.5,-56 2715.5,-62 2709.5,-68 2703.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"2602\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.157</text>\r\n<text text-anchor=\"start\" x=\"2596\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 35</text>\r\n<text text-anchor=\"start\" x=\"2595\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [3, 32]</text>\r\n<text text-anchor=\"start\" x=\"2571.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 29&#45;&gt;31 -->\r\n<g id=\"edge31\" class=\"edge\"><title>29&#45;&gt;31</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2532.17,-103.726C2547.49,-93.9161 2563.82,-83.4644 2578.99,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2581.24,-76.4654 2587.77,-68.1252 2577.46,-70.5704 2581.24,-76.4654\"/>\r\n</g>\r\n<!-- 33 -->\r\n<g id=\"node34\" class=\"node\"><title>33</title>\r\n<path fill=\"#e9965a\" stroke=\"black\" d=\"M3473,-425C3473,-425 3344,-425 3344,-425 3338,-425 3332,-419 3332,-413 3332,-413 3332,-354 3332,-354 3332,-348 3338,-342 3344,-342 3344,-342 3473,-342 3473,-342 3479,-342 3485,-348 3485,-354 3485,-354 3485,-413 3485,-413 3485,-419 3479,-425 3473,-425\"/>\r\n<text text-anchor=\"start\" x=\"3352.5\" y=\"-409.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_y ≤ 70.8</text>\r\n<text text-anchor=\"start\" x=\"3371\" y=\"-394.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.248</text>\r\n<text text-anchor=\"start\" x=\"3361\" y=\"-379.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 311</text>\r\n<text text-anchor=\"start\" x=\"3355.5\" y=\"-364.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [266, 45]</text>\r\n<text text-anchor=\"start\" x=\"3340\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 32&#45;&gt;33 -->\r\n<g id=\"edge33\" class=\"edge\"><title>32&#45;&gt;33</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3408.5,-460.907C3408.5,-452.649 3408.5,-443.864 3408.5,-435.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3412,-435.021 3408.5,-425.021 3405,-435.021 3412,-435.021\"/>\r\n</g>\r\n<!-- 48 -->\r\n<g id=\"node49\" class=\"node\"><title>48</title>\r\n<path fill=\"#fdf5f0\" stroke=\"black\" d=\"M4498,-425C4498,-425 4369,-425 4369,-425 4363,-425 4357,-419 4357,-413 4357,-413 4357,-354 4357,-354 4357,-348 4363,-342 4369,-342 4369,-342 4498,-342 4498,-342 4504,-342 4510,-348 4510,-354 4510,-354 4510,-413 4510,-413 4510,-419 4504,-425 4498,-425\"/>\r\n<text text-anchor=\"start\" x=\"4367\" y=\"-409.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Towers lost_x ≤ 4.15</text>\r\n<text text-anchor=\"start\" x=\"4396\" y=\"-394.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.499</text>\r\n<text text-anchor=\"start\" x=\"4386\" y=\"-379.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 331</text>\r\n<text text-anchor=\"start\" x=\"4376.5\" y=\"-364.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [172, 159]</text>\r\n<text text-anchor=\"start\" x=\"4365\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 32&#45;&gt;48 -->\r\n<g id=\"edge48\" class=\"edge\"><title>32&#45;&gt;48</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3487.65,-492.465C3676.2,-470.943 4150.71,-416.779 4346.72,-394.406\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4347.15,-397.879 4356.69,-393.268 4346.36,-390.924 4347.15,-397.879\"/>\r\n</g>\r\n<!-- 34 -->\r\n<g id=\"node35\" class=\"node\"><title>34</title>\r\n<path fill=\"#eb9e67\" stroke=\"black\" d=\"M3302,-306C3302,-306 3173,-306 3173,-306 3167,-306 3161,-300 3161,-294 3161,-294 3161,-235 3161,-235 3161,-229 3167,-223 3173,-223 3173,-223 3302,-223 3302,-223 3308,-223 3314,-229 3314,-235 3314,-235 3314,-294 3314,-294 3314,-300 3308,-306 3302,-306\"/>\r\n<text text-anchor=\"start\" x=\"3190.5\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">WPM_x ≤ 3.25</text>\r\n<text text-anchor=\"start\" x=\"3200\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.306</text>\r\n<text text-anchor=\"start\" x=\"3190\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 223</text>\r\n<text text-anchor=\"start\" x=\"3184.5\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [181, 42]</text>\r\n<text text-anchor=\"start\" x=\"3169\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 33&#45;&gt;34 -->\r\n<g id=\"edge34\" class=\"edge\"><title>33&#45;&gt;34</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3349.17,-341.907C3334.98,-332.197 3319.72,-321.758 3305.19,-311.811\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3306.95,-308.779 3296.72,-306.021 3303,-314.556 3306.95,-308.779\"/>\r\n</g>\r\n<!-- 41 -->\r\n<g id=\"node42\" class=\"node\"><title>41</title>\r\n<path fill=\"#e68540\" stroke=\"black\" d=\"M3644,-306C3644,-306 3515,-306 3515,-306 3509,-306 3503,-300 3503,-294 3503,-294 3503,-235 3503,-235 3503,-229 3509,-223 3515,-223 3515,-223 3644,-223 3644,-223 3650,-223 3656,-229 3656,-235 3656,-235 3656,-294 3656,-294 3656,-300 3650,-306 3644,-306\"/>\r\n<text text-anchor=\"start\" x=\"3519\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">DRA@15_x ≤ 1.12</text>\r\n<text text-anchor=\"start\" x=\"3542\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.066</text>\r\n<text text-anchor=\"start\" x=\"3536\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 88</text>\r\n<text text-anchor=\"start\" x=\"3535\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [85, 3]</text>\r\n<text text-anchor=\"start\" x=\"3511\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 33&#45;&gt;41 -->\r\n<g id=\"edge41\" class=\"edge\"><title>33&#45;&gt;41</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3467.83,-341.907C3482.02,-332.197 3497.28,-321.758 3511.81,-311.811\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3514,-314.556 3520.28,-306.021 3510.05,-308.779 3514,-314.556\"/>\r\n</g>\r\n<!-- 35 -->\r\n<g id=\"node36\" class=\"node\"><title>35</title>\r\n<path fill=\"#f0b68d\" stroke=\"black\" d=\"M3045,-187C3045,-187 2916,-187 2916,-187 2910,-187 2904,-181 2904,-175 2904,-175 2904,-116 2904,-116 2904,-110 2910,-104 2916,-104 2916,-104 3045,-104 3045,-104 3051,-104 3057,-110 3057,-116 3057,-116 3057,-175 3057,-175 3057,-181 3051,-187 3045,-187\"/>\r\n<text text-anchor=\"start\" x=\"2924\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">WCPM_x ≤ 1.265</text>\r\n<text text-anchor=\"start\" x=\"2943\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.417</text>\r\n<text text-anchor=\"start\" x=\"2937\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 64</text>\r\n<text text-anchor=\"start\" x=\"2932\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [45, 19]</text>\r\n<text text-anchor=\"start\" x=\"2912\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 34&#45;&gt;35 -->\r\n<g id=\"edge35\" class=\"edge\"><title>34&#45;&gt;35</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3160.74,-228.555C3131.12,-215.069 3097.09,-199.577 3066.64,-185.714\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3067.86,-182.425 3057.31,-181.467 3064.96,-188.796 3067.86,-182.425\"/>\r\n</g>\r\n<!-- 38 -->\r\n<g id=\"node39\" class=\"node\"><title>38</title>\r\n<path fill=\"#e9965a\" stroke=\"black\" d=\"M3302,-187C3302,-187 3173,-187 3173,-187 3167,-187 3161,-181 3161,-175 3161,-175 3161,-116 3161,-116 3161,-110 3167,-104 3173,-104 3173,-104 3302,-104 3302,-104 3308,-104 3314,-110 3314,-116 3314,-116 3314,-175 3314,-175 3314,-181 3308,-187 3302,-187\"/>\r\n<text text-anchor=\"start\" x=\"3181.5\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VWPM_x ≤ 1.655</text>\r\n<text text-anchor=\"start\" x=\"3200\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.247</text>\r\n<text text-anchor=\"start\" x=\"3190\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 159</text>\r\n<text text-anchor=\"start\" x=\"3184.5\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [136, 23]</text>\r\n<text text-anchor=\"start\" x=\"3169\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 34&#45;&gt;38 -->\r\n<g id=\"edge38\" class=\"edge\"><title>34&#45;&gt;38</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3237.5,-222.907C3237.5,-214.649 3237.5,-205.864 3237.5,-197.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3241,-197.021 3237.5,-187.021 3234,-197.021 3241,-197.021\"/>\r\n</g>\r\n<!-- 36 -->\r\n<g id=\"node37\" class=\"node\"><title>36</title>\r\n<path fill=\"#6ab6ec\" stroke=\"black\" d=\"M2873.5,-68C2873.5,-68 2745.5,-68 2745.5,-68 2739.5,-68 2733.5,-62 2733.5,-56 2733.5,-56 2733.5,-12 2733.5,-12 2733.5,-6 2739.5,-0 2745.5,-0 2745.5,-0 2873.5,-0 2873.5,-0 2879.5,-0 2885.5,-6 2885.5,-12 2885.5,-12 2885.5,-56 2885.5,-56 2885.5,-62 2879.5,-68 2873.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"2776\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.32</text>\r\n<text text-anchor=\"start\" x=\"2770\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 5</text>\r\n<text text-anchor=\"start\" x=\"2769\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [1, 4]</text>\r\n<text text-anchor=\"start\" x=\"2741.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 35&#45;&gt;36 -->\r\n<g id=\"edge36\" class=\"edge\"><title>35&#45;&gt;36</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2916.83,-103.726C2901.51,-93.9161 2885.18,-83.4644 2870.01,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2871.54,-70.5704 2861.23,-68.1252 2867.76,-76.4654 2871.54,-70.5704\"/>\r\n</g>\r\n<!-- 37 -->\r\n<g id=\"node38\" class=\"node\"><title>37</title>\r\n<path fill=\"#eeac7c\" stroke=\"black\" d=\"M3045,-68C3045,-68 2916,-68 2916,-68 2910,-68 2904,-62 2904,-56 2904,-56 2904,-12 2904,-12 2904,-6 2910,-0 2916,-0 2916,-0 3045,-0 3045,-0 3051,-0 3057,-6 3057,-12 3057,-12 3057,-56 3057,-56 3057,-62 3051,-68 3045,-68\"/>\r\n<text text-anchor=\"start\" x=\"2943\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.379</text>\r\n<text text-anchor=\"start\" x=\"2937\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 59</text>\r\n<text text-anchor=\"start\" x=\"2932\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [44, 15]</text>\r\n<text text-anchor=\"start\" x=\"2912\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 35&#45;&gt;37 -->\r\n<g id=\"edge37\" class=\"edge\"><title>35&#45;&gt;37</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M2980.5,-103.726C2980.5,-95.5175 2980.5,-86.8595 2980.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"2984,-78.2996 2980.5,-68.2996 2977,-78.2996 2984,-78.2996\"/>\r\n</g>\r\n<!-- 39 -->\r\n<g id=\"node40\" class=\"node\"><title>39</title>\r\n<path fill=\"#e99254\" stroke=\"black\" d=\"M3216,-68C3216,-68 3087,-68 3087,-68 3081,-68 3075,-62 3075,-56 3075,-56 3075,-12 3075,-12 3075,-6 3081,-0 3087,-0 3087,-0 3216,-0 3216,-0 3222,-0 3228,-6 3228,-12 3228,-12 3228,-56 3228,-56 3228,-62 3222,-68 3216,-68\"/>\r\n<text text-anchor=\"start\" x=\"3114\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.214</text>\r\n<text text-anchor=\"start\" x=\"3104\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 148</text>\r\n<text text-anchor=\"start\" x=\"3098.5\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [130, 18]</text>\r\n<text text-anchor=\"start\" x=\"3083\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 38&#45;&gt;39 -->\r\n<g id=\"edge39\" class=\"edge\"><title>38&#45;&gt;39</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3205.48,-103.726C3198.46,-94.7878 3191.02,-85.3168 3183.98,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3186.58,-74.0022 3177.65,-68.2996 3181.08,-78.3259 3186.58,-74.0022\"/>\r\n</g>\r\n<!-- 40 -->\r\n<g id=\"node41\" class=\"node\"><title>40</title>\r\n<path fill=\"#fbeade\" stroke=\"black\" d=\"M3387,-68C3387,-68 3258,-68 3258,-68 3252,-68 3246,-62 3246,-56 3246,-56 3246,-12 3246,-12 3246,-6 3252,-0 3258,-0 3258,-0 3387,-0 3387,-0 3393,-0 3399,-6 3399,-12 3399,-12 3399,-56 3399,-56 3399,-62 3393,-68 3387,-68\"/>\r\n<text text-anchor=\"start\" x=\"3285\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.496</text>\r\n<text text-anchor=\"start\" x=\"3279\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 11</text>\r\n<text text-anchor=\"start\" x=\"3282\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [6, 5]</text>\r\n<text text-anchor=\"start\" x=\"3254\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 38&#45;&gt;40 -->\r\n<g id=\"edge40\" class=\"edge\"><title>38&#45;&gt;40</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3269.15,-103.726C3276.09,-94.7878 3283.44,-85.3168 3290.4,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3293.28,-78.3451 3296.65,-68.2996 3287.75,-74.0528 3293.28,-78.3451\"/>\r\n</g>\r\n<!-- 42 -->\r\n<g id=\"node43\" class=\"node\"><title>42</title>\r\n<path fill=\"#e5833b\" stroke=\"black\" d=\"M3644,-187C3644,-187 3515,-187 3515,-187 3509,-187 3503,-181 3503,-175 3503,-175 3503,-116 3503,-116 3503,-110 3509,-104 3515,-104 3515,-104 3644,-104 3644,-104 3650,-104 3656,-110 3656,-116 3656,-116 3656,-175 3656,-175 3656,-181 3650,-187 3644,-187\"/>\r\n<text text-anchor=\"start\" x=\"3530.5\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">FT%_x ≤ 11.95</text>\r\n<text text-anchor=\"start\" x=\"3542\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.024</text>\r\n<text text-anchor=\"start\" x=\"3536\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 83</text>\r\n<text text-anchor=\"start\" x=\"3535\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [82, 1]</text>\r\n<text text-anchor=\"start\" x=\"3511\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 41&#45;&gt;42 -->\r\n<g id=\"edge42\" class=\"edge\"><title>41&#45;&gt;42</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3579.5,-222.907C3579.5,-214.649 3579.5,-205.864 3579.5,-197.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3583,-197.021 3579.5,-187.021 3576,-197.021 3583,-197.021\"/>\r\n</g>\r\n<!-- 45 -->\r\n<g id=\"node46\" class=\"node\"><title>45</title>\r\n<path fill=\"#f6d5bd\" stroke=\"black\" d=\"M3900,-187C3900,-187 3771,-187 3771,-187 3765,-187 3759,-181 3759,-175 3759,-175 3759,-116 3759,-116 3759,-110 3765,-104 3771,-104 3771,-104 3900,-104 3900,-104 3906,-104 3912,-110 3912,-116 3912,-116 3912,-175 3912,-175 3912,-181 3906,-187 3900,-187\"/>\r\n<text text-anchor=\"start\" x=\"3784\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">DRA%_y ≤ 64.8</text>\r\n<text text-anchor=\"start\" x=\"3802\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.48</text>\r\n<text text-anchor=\"start\" x=\"3796\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 5</text>\r\n<text text-anchor=\"start\" x=\"3795\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [3, 2]</text>\r\n<text text-anchor=\"start\" x=\"3767\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 41&#45;&gt;45 -->\r\n<g id=\"edge45\" class=\"edge\"><title>41&#45;&gt;45</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3656.31,-228.396C3685.67,-214.976 3719.34,-199.589 3749.5,-185.802\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3751.11,-188.919 3758.75,-181.579 3748.2,-182.553 3751.11,-188.919\"/>\r\n</g>\r\n<!-- 43 -->\r\n<g id=\"node44\" class=\"node\"><title>43</title>\r\n<path fill=\"#f2c09c\" stroke=\"black\" d=\"M3558,-68C3558,-68 3429,-68 3429,-68 3423,-68 3417,-62 3417,-56 3417,-56 3417,-12 3417,-12 3417,-6 3423,-0 3429,-0 3429,-0 3558,-0 3558,-0 3564,-0 3570,-6 3570,-12 3570,-12 3570,-56 3570,-56 3570,-62 3564,-68 3558,-68\"/>\r\n<text text-anchor=\"start\" x=\"3456\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.444</text>\r\n<text text-anchor=\"start\" x=\"3454\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 3</text>\r\n<text text-anchor=\"start\" x=\"3453\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [2, 1]</text>\r\n<text text-anchor=\"start\" x=\"3425\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 42&#45;&gt;43 -->\r\n<g id=\"edge43\" class=\"edge\"><title>42&#45;&gt;43</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3547.48,-103.726C3540.46,-94.7878 3533.02,-85.3168 3525.98,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3528.58,-74.0022 3519.65,-68.2996 3523.08,-78.3259 3528.58,-74.0022\"/>\r\n</g>\r\n<!-- 44 -->\r\n<g id=\"node45\" class=\"node\"><title>44</title>\r\n<path fill=\"#e58139\" stroke=\"black\" d=\"M3729,-68C3729,-68 3600,-68 3600,-68 3594,-68 3588,-62 3588,-56 3588,-56 3588,-12 3588,-12 3588,-6 3594,-0 3600,-0 3600,-0 3729,-0 3729,-0 3735,-0 3741,-6 3741,-12 3741,-12 3741,-56 3741,-56 3741,-62 3735,-68 3729,-68\"/>\r\n<text text-anchor=\"start\" x=\"3635.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"3621\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 80</text>\r\n<text text-anchor=\"start\" x=\"3620\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [80, 0]</text>\r\n<text text-anchor=\"start\" x=\"3596\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 42&#45;&gt;44 -->\r\n<g id=\"edge44\" class=\"edge\"><title>42&#45;&gt;44</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3611.15,-103.726C3618.09,-94.7878 3625.44,-85.3168 3632.4,-76.3558\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3635.28,-78.3451 3638.65,-68.2996 3629.75,-74.0528 3635.28,-78.3451\"/>\r\n</g>\r\n<!-- 46 -->\r\n<g id=\"node47\" class=\"node\"><title>46</title>\r\n<path fill=\"#e58139\" stroke=\"black\" d=\"M3900,-68C3900,-68 3771,-68 3771,-68 3765,-68 3759,-62 3759,-56 3759,-56 3759,-12 3759,-12 3759,-6 3765,-0 3771,-0 3771,-0 3900,-0 3900,-0 3906,-0 3912,-6 3912,-12 3912,-12 3912,-56 3912,-56 3912,-62 3906,-68 3900,-68\"/>\r\n<text text-anchor=\"start\" x=\"3806.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"3796\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 3</text>\r\n<text text-anchor=\"start\" x=\"3795\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [3, 0]</text>\r\n<text text-anchor=\"start\" x=\"3767\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 45&#45;&gt;46 -->\r\n<g id=\"edge46\" class=\"edge\"><title>45&#45;&gt;46</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3835.5,-103.726C3835.5,-95.5175 3835.5,-86.8595 3835.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3839,-78.2996 3835.5,-68.2996 3832,-78.2996 3839,-78.2996\"/>\r\n</g>\r\n<!-- 47 -->\r\n<g id=\"node48\" class=\"node\"><title>47</title>\r\n<path fill=\"#399de5\" stroke=\"black\" d=\"M4070.5,-68C4070.5,-68 3942.5,-68 3942.5,-68 3936.5,-68 3930.5,-62 3930.5,-56 3930.5,-56 3930.5,-12 3930.5,-12 3930.5,-6 3936.5,-0 3942.5,-0 3942.5,-0 4070.5,-0 4070.5,-0 4076.5,-0 4082.5,-6 4082.5,-12 4082.5,-12 4082.5,-56 4082.5,-56 4082.5,-62 4076.5,-68 4070.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"3977.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"3967\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 2</text>\r\n<text text-anchor=\"start\" x=\"3966\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [0, 2]</text>\r\n<text text-anchor=\"start\" x=\"3938.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 45&#45;&gt;47 -->\r\n<g id=\"edge47\" class=\"edge\"><title>45&#45;&gt;47</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M3899.17,-103.726C3914.49,-93.9161 3930.82,-83.4644 3945.99,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"3948.24,-76.4654 3954.77,-68.1252 3944.46,-70.5704 3948.24,-76.4654\"/>\r\n</g>\r\n<!-- 49 -->\r\n<g id=\"node50\" class=\"node\"><title>49</title>\r\n<path fill=\"#87c4ef\" stroke=\"black\" d=\"M4497.5,-306C4497.5,-306 4369.5,-306 4369.5,-306 4363.5,-306 4357.5,-300 4357.5,-294 4357.5,-294 4357.5,-235 4357.5,-235 4357.5,-229 4363.5,-223 4369.5,-223 4369.5,-223 4497.5,-223 4497.5,-223 4503.5,-223 4509.5,-229 4509.5,-235 4509.5,-235 4509.5,-294 4509.5,-294 4509.5,-300 4503.5,-306 4497.5,-306\"/>\r\n<text text-anchor=\"start\" x=\"4377.5\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">VWPM_y ≤ 1.295</text>\r\n<text text-anchor=\"start\" x=\"4396\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.406</text>\r\n<text text-anchor=\"start\" x=\"4390\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 53</text>\r\n<text text-anchor=\"start\" x=\"4385\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [15, 38]</text>\r\n<text text-anchor=\"start\" x=\"4365.5\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 48&#45;&gt;49 -->\r\n<g id=\"edge49\" class=\"edge\"><title>48&#45;&gt;49</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4433.5,-341.907C4433.5,-333.649 4433.5,-324.864 4433.5,-316.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4437,-316.021 4433.5,-306.021 4430,-316.021 4437,-316.021\"/>\r\n</g>\r\n<!-- 56 -->\r\n<g id=\"node57\" class=\"node\"><title>56</title>\r\n<path fill=\"#f9e2d2\" stroke=\"black\" d=\"M4994,-306C4994,-306 4865,-306 4865,-306 4859,-306 4853,-300 4853,-294 4853,-294 4853,-235 4853,-235 4853,-229 4859,-223 4865,-223 4865,-223 4994,-223 4994,-223 5000,-223 5006,-229 5006,-235 5006,-235 5006,-294 5006,-294 5006,-300 5000,-306 4994,-306\"/>\r\n<text text-anchor=\"start\" x=\"4873.5\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Win rate_y ≤ 58.9</text>\r\n<text text-anchor=\"start\" x=\"4892\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.492</text>\r\n<text text-anchor=\"start\" x=\"4882\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 278</text>\r\n<text text-anchor=\"start\" x=\"4872.5\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [157, 121]</text>\r\n<text text-anchor=\"start\" x=\"4861\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 48&#45;&gt;56 -->\r\n<g id=\"edge56\" class=\"edge\"><title>48&#45;&gt;56</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4510.18,-364.411C4600.13,-343.193 4748.89,-308.105 4842.91,-285.925\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4843.75,-289.323 4852.68,-283.62 4842.15,-282.51 4843.75,-289.323\"/>\r\n</g>\r\n<!-- 50 -->\r\n<g id=\"node51\" class=\"node\"><title>50</title>\r\n<path fill=\"#d7ebfa\" stroke=\"black\" d=\"M4414,-187C4414,-187 4281,-187 4281,-187 4275,-187 4269,-181 4269,-175 4269,-175 4269,-116 4269,-116 4269,-110 4275,-104 4281,-104 4281,-104 4414,-104 4414,-104 4420,-104 4426,-110 4426,-116 4426,-116 4426,-175 4426,-175 4426,-181 4420,-187 4414,-187\"/>\r\n<text text-anchor=\"start\" x=\"4277\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Kills / game_x ≤ 16.55</text>\r\n<text text-anchor=\"start\" x=\"4310\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.494</text>\r\n<text text-anchor=\"start\" x=\"4304\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 27</text>\r\n<text text-anchor=\"start\" x=\"4299\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [12, 15]</text>\r\n<text text-anchor=\"start\" x=\"4279.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 49&#45;&gt;50 -->\r\n<g id=\"edge50\" class=\"edge\"><title>49&#45;&gt;50</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4403.66,-222.907C4397.13,-214.014 4390.14,-204.509 4383.39,-195.331\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4386.03,-193.005 4377.28,-187.021 4380.39,-197.151 4386.03,-193.005\"/>\r\n</g>\r\n<!-- 53 -->\r\n<g id=\"node54\" class=\"node\"><title>53</title>\r\n<path fill=\"#53aae8\" stroke=\"black\" d=\"M4584.5,-187C4584.5,-187 4456.5,-187 4456.5,-187 4450.5,-187 4444.5,-181 4444.5,-175 4444.5,-175 4444.5,-116 4444.5,-116 4444.5,-110 4450.5,-104 4456.5,-104 4456.5,-104 4584.5,-104 4584.5,-104 4590.5,-104 4596.5,-110 4596.5,-116 4596.5,-116 4596.5,-175 4596.5,-175 4596.5,-181 4590.5,-187 4584.5,-187\"/>\r\n<text text-anchor=\"start\" x=\"4469\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">DRA%_y ≤ 50.2</text>\r\n<text text-anchor=\"start\" x=\"4483\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.204</text>\r\n<text text-anchor=\"start\" x=\"4477\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 26</text>\r\n<text text-anchor=\"start\" x=\"4476\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [3, 23]</text>\r\n<text text-anchor=\"start\" x=\"4452.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 49&#45;&gt;53 -->\r\n<g id=\"edge53\" class=\"edge\"><title>49&#45;&gt;53</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4463.68,-222.907C4470.36,-213.923 4477.51,-204.315 4484.4,-195.05\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4487.21,-197.134 4490.37,-187.021 4481.59,-192.957 4487.21,-197.134\"/>\r\n</g>\r\n<!-- 51 -->\r\n<g id=\"node52\" class=\"node\"><title>51</title>\r\n<path fill=\"#f8ddc9\" stroke=\"black\" d=\"M4242,-68C4242,-68 4113,-68 4113,-68 4107,-68 4101,-62 4101,-56 4101,-56 4101,-12 4101,-12 4101,-6 4107,-0 4113,-0 4113,-0 4242,-0 4242,-0 4248,-0 4254,-6 4254,-12 4254,-12 4254,-56 4254,-56 4254,-62 4248,-68 4242,-68\"/>\r\n<text text-anchor=\"start\" x=\"4140\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.488</text>\r\n<text text-anchor=\"start\" x=\"4134\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 19</text>\r\n<text text-anchor=\"start\" x=\"4133\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [11, 8]</text>\r\n<text text-anchor=\"start\" x=\"4109\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 50&#45;&gt;51 -->\r\n<g id=\"edge51\" class=\"edge\"><title>50&#45;&gt;51</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4284.2,-103.726C4268.97,-93.9161 4252.74,-83.4644 4237.66,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4239.23,-70.5979 4228.93,-68.1252 4235.44,-76.4828 4239.23,-70.5979\"/>\r\n</g>\r\n<!-- 52 -->\r\n<g id=\"node53\" class=\"node\"><title>52</title>\r\n<path fill=\"#55abe9\" stroke=\"black\" d=\"M4412.5,-68C4412.5,-68 4284.5,-68 4284.5,-68 4278.5,-68 4272.5,-62 4272.5,-56 4272.5,-56 4272.5,-12 4272.5,-12 4272.5,-6 4278.5,-0 4284.5,-0 4284.5,-0 4412.5,-0 4412.5,-0 4418.5,-0 4424.5,-6 4424.5,-12 4424.5,-12 4424.5,-56 4424.5,-56 4424.5,-62 4418.5,-68 4412.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"4311\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.219</text>\r\n<text text-anchor=\"start\" x=\"4309\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 8</text>\r\n<text text-anchor=\"start\" x=\"4308\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [1, 7]</text>\r\n<text text-anchor=\"start\" x=\"4280.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 50&#45;&gt;52 -->\r\n<g id=\"edge52\" class=\"edge\"><title>50&#45;&gt;52</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4347.87,-103.726C4347.95,-95.5175 4348.03,-86.8595 4348.1,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4351.6,-78.3312 4348.2,-68.2996 4344.6,-78.2672 4351.6,-78.3312\"/>\r\n</g>\r\n<!-- 54 -->\r\n<g id=\"node55\" class=\"node\"><title>54</title>\r\n<path fill=\"#e58139\" stroke=\"black\" d=\"M4584,-68C4584,-68 4455,-68 4455,-68 4449,-68 4443,-62 4443,-56 4443,-56 4443,-12 4443,-12 4443,-6 4449,-0 4455,-0 4455,-0 4584,-0 4584,-0 4590,-0 4596,-6 4596,-12 4596,-12 4596,-56 4596,-56 4596,-62 4590,-68 4584,-68\"/>\r\n<text text-anchor=\"start\" x=\"4490.5\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.0</text>\r\n<text text-anchor=\"start\" x=\"4480\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 1</text>\r\n<text text-anchor=\"start\" x=\"4479\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [1, 0]</text>\r\n<text text-anchor=\"start\" x=\"4451\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 53&#45;&gt;54 -->\r\n<g id=\"edge54\" class=\"edge\"><title>53&#45;&gt;54</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4520.13,-103.726C4520.05,-95.5175 4519.97,-86.8595 4519.9,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4523.4,-78.2672 4519.8,-68.2996 4516.4,-78.3312 4523.4,-78.2672\"/>\r\n</g>\r\n<!-- 55 -->\r\n<g id=\"node56\" class=\"node\"><title>55</title>\r\n<path fill=\"#4aa6e7\" stroke=\"black\" d=\"M4754.5,-68C4754.5,-68 4626.5,-68 4626.5,-68 4620.5,-68 4614.5,-62 4614.5,-56 4614.5,-56 4614.5,-12 4614.5,-12 4614.5,-6 4620.5,-0 4626.5,-0 4626.5,-0 4754.5,-0 4754.5,-0 4760.5,-0 4766.5,-6 4766.5,-12 4766.5,-12 4766.5,-56 4766.5,-56 4766.5,-62 4760.5,-68 4754.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"4653\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.147</text>\r\n<text text-anchor=\"start\" x=\"4647\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 25</text>\r\n<text text-anchor=\"start\" x=\"4646\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [2, 23]</text>\r\n<text text-anchor=\"start\" x=\"4622.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 53&#45;&gt;55 -->\r\n<g id=\"edge55\" class=\"edge\"><title>53&#45;&gt;55</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4583.8,-103.726C4599.03,-93.9161 4615.26,-83.4644 4630.34,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4632.56,-76.4828 4639.07,-68.1252 4628.77,-70.5979 4632.56,-76.4828\"/>\r\n</g>\r\n<!-- 57 -->\r\n<g id=\"node58\" class=\"node\"><title>57</title>\r\n<path fill=\"#b5daf5\" stroke=\"black\" d=\"M5010,-187C5010,-187 4849,-187 4849,-187 4843,-187 4837,-181 4837,-175 4837,-175 4837,-116 4837,-116 4837,-110 4843,-104 4849,-104 4849,-104 5010,-104 5010,-104 5016,-104 5022,-110 5022,-116 5022,-116 5022,-175 5022,-175 5022,-181 5016,-187 5010,-187\"/>\r\n<text text-anchor=\"start\" x=\"4845\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Game duration_x ≤ 2035.5</text>\r\n<text text-anchor=\"start\" x=\"4892\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.474</text>\r\n<text text-anchor=\"start\" x=\"4886\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 83</text>\r\n<text text-anchor=\"start\" x=\"4881\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [32, 51]</text>\r\n<text text-anchor=\"start\" x=\"4861.5\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 56&#45;&gt;57 -->\r\n<g id=\"edge57\" class=\"edge\"><title>56&#45;&gt;57</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4929.5,-222.907C4929.5,-214.649 4929.5,-205.864 4929.5,-197.302\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4933,-197.021 4929.5,-187.021 4926,-197.021 4933,-197.021\"/>\r\n</g>\r\n<!-- 60 -->\r\n<g id=\"node61\" class=\"node\"><title>60</title>\r\n<path fill=\"#f4c8a8\" stroke=\"black\" d=\"M5283,-187C5283,-187 5122,-187 5122,-187 5116,-187 5110,-181 5110,-175 5110,-175 5110,-116 5110,-116 5110,-110 5116,-104 5122,-104 5122,-104 5283,-104 5283,-104 5289,-104 5295,-110 5295,-116 5295,-116 5295,-175 5295,-175 5295,-181 5289,-187 5283,-187\"/>\r\n<text text-anchor=\"start\" x=\"5118\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Game duration_x ≤ 2023.5</text>\r\n<text text-anchor=\"start\" x=\"5169\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.46</text>\r\n<text text-anchor=\"start\" x=\"5155\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 195</text>\r\n<text text-anchor=\"start\" x=\"5149.5\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [125, 70]</text>\r\n<text text-anchor=\"start\" x=\"5134\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 56&#45;&gt;60 -->\r\n<g id=\"edge60\" class=\"edge\"><title>56&#45;&gt;60</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M5006.27,-230.598C5035.43,-218.102 5069.11,-203.666 5100.26,-190.319\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"5102.02,-193.372 5109.83,-186.215 5099.26,-186.938 5102.02,-193.372\"/>\r\n</g>\r\n<!-- 58 -->\r\n<g id=\"node59\" class=\"node\"><title>58</title>\r\n<path fill=\"#85c3ef\" stroke=\"black\" d=\"M4924.5,-68C4924.5,-68 4796.5,-68 4796.5,-68 4790.5,-68 4784.5,-62 4784.5,-56 4784.5,-56 4784.5,-12 4784.5,-12 4784.5,-6 4790.5,-0 4796.5,-0 4796.5,-0 4924.5,-0 4924.5,-0 4930.5,-0 4936.5,-6 4936.5,-12 4936.5,-12 4936.5,-56 4936.5,-56 4936.5,-62 4930.5,-68 4924.5,-68\"/>\r\n<text text-anchor=\"start\" x=\"4823\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.401</text>\r\n<text text-anchor=\"start\" x=\"4817\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 54</text>\r\n<text text-anchor=\"start\" x=\"4812\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [15, 39]</text>\r\n<text text-anchor=\"start\" x=\"4792.5\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Red team win</text>\r\n</g>\r\n<!-- 57&#45;&gt;58 -->\r\n<g id=\"edge58\" class=\"edge\"><title>57&#45;&gt;58</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4903.81,-103.726C4898.29,-94.9703 4892.45,-85.7032 4886.91,-76.9051\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4889.78,-74.8941 4881.48,-68.2996 4883.85,-78.6259 4889.78,-74.8941\"/>\r\n</g>\r\n<!-- 59 -->\r\n<g id=\"node60\" class=\"node\"><title>59</title>\r\n<path fill=\"#f7dac5\" stroke=\"black\" d=\"M5096,-68C5096,-68 4967,-68 4967,-68 4961,-68 4955,-62 4955,-56 4955,-56 4955,-12 4955,-12 4955,-6 4961,-0 4967,-0 4967,-0 5096,-0 5096,-0 5102,-0 5108,-6 5108,-12 5108,-12 5108,-56 5108,-56 5108,-62 5102,-68 5096,-68\"/>\r\n<text text-anchor=\"start\" x=\"4994\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.485</text>\r\n<text text-anchor=\"start\" x=\"4988\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 29</text>\r\n<text text-anchor=\"start\" x=\"4983\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [17, 12]</text>\r\n<text text-anchor=\"start\" x=\"4963\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 57&#45;&gt;59 -->\r\n<g id=\"edge59\" class=\"edge\"><title>57&#45;&gt;59</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M4967.48,-103.726C4975.98,-94.6054 4984.99,-84.93 4993.49,-75.8078\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"4996.23,-78.0025 5000.48,-68.2996 4991.1,-73.2312 4996.23,-78.0025\"/>\r\n</g>\r\n<!-- 61 -->\r\n<g id=\"node62\" class=\"node\"><title>61</title>\r\n<path fill=\"#f0b68d\" stroke=\"black\" d=\"M5267,-68C5267,-68 5138,-68 5138,-68 5132,-68 5126,-62 5126,-56 5126,-56 5126,-12 5126,-12 5126,-6 5132,-0 5138,-0 5138,-0 5267,-0 5267,-0 5273,-0 5279,-6 5279,-12 5279,-12 5279,-56 5279,-56 5279,-62 5273,-68 5267,-68\"/>\r\n<text text-anchor=\"start\" x=\"5165\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.417</text>\r\n<text text-anchor=\"start\" x=\"5155\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 128</text>\r\n<text text-anchor=\"start\" x=\"5154\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [90, 38]</text>\r\n<text text-anchor=\"start\" x=\"5134\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 60&#45;&gt;61 -->\r\n<g id=\"edge61\" class=\"edge\"><title>60&#45;&gt;61</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M5202.5,-103.726C5202.5,-95.5175 5202.5,-86.8595 5202.5,-78.56\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"5206,-78.2996 5202.5,-68.2996 5199,-78.2996 5206,-78.2996\"/>\r\n</g>\r\n<!-- 62 -->\r\n<g id=\"node63\" class=\"node\"><title>62</title>\r\n<path fill=\"#fdf4ee\" stroke=\"black\" d=\"M5438,-68C5438,-68 5309,-68 5309,-68 5303,-68 5297,-62 5297,-56 5297,-56 5297,-12 5297,-12 5297,-6 5303,-0 5309,-0 5309,-0 5438,-0 5438,-0 5444,-0 5450,-6 5450,-12 5450,-12 5450,-56 5450,-56 5450,-62 5444,-68 5438,-68\"/>\r\n<text text-anchor=\"start\" x=\"5336\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">gini = 0.499</text>\r\n<text text-anchor=\"start\" x=\"5330\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">samples = 67</text>\r\n<text text-anchor=\"start\" x=\"5325\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">value = [35, 32]</text>\r\n<text text-anchor=\"start\" x=\"5305\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">class = Blue team win</text>\r\n</g>\r\n<!-- 60&#45;&gt;62 -->\r\n<g id=\"edge62\" class=\"edge\"><title>60&#45;&gt;62</title>\r\n<path fill=\"none\" stroke=\"black\" d=\"M5266.17,-103.726C5281.49,-93.9161 5297.82,-83.4644 5312.99,-73.7496\"/>\r\n<polygon fill=\"black\" stroke=\"black\" points=\"5315.24,-76.4654 5321.77,-68.1252 5311.46,-70.5704 5315.24,-76.4654\"/>\r\n</g>\r\n</g>\r\n</svg>\r\n",
      "text/plain": [
       "<graphviz.files.Source at 0x2ccd047f5e0>"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "colnames = final.columns\n",
    "colnames = colnames.drop(\"Score\")\n",
    "\n",
    "classifier = tree.DecisionTreeClassifier(max_depth=5)\n",
    "classifier.fit(data, onedLabels)\n",
    "\n",
    "dot_data = tree.export_graphviz(\n",
    "    classifier,\n",
    "    out_file=None,\n",
    "    feature_names=colnames,\n",
    "    class_names=['Blue team win','Red team win'],\n",
    "    filled=True,\n",
    "    rounded=True,\n",
    "    special_characters=True\n",
    ")\n",
    "\n",
    "# Visualize the decision tree\n",
    "pydot_graph = pydotplus.graph_from_dot_data(dot_data)\n",
    "pydot_graph.set_size(5,5)\n",
    "pydot_graph.write_png('decisionTree.png')\n",
    "graph = graphviz.Source(dot_data)\n",
    "graph"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we can see interesting observations about the most significant factors to determine winner. The biggest determinator is of course **Win rate_x**, blue team's winrate, which cuts the whole dataset into half. After the most self-explanatory factors, winrates, we get to see more interesting observations on depth 3.\n",
    "\n",
    "There we have some of the most important factors to determine a winner: **Towers lost**, **Game duration** (average of the whole split) and **DPM**. Towers and damage are of course somewhat self-explanatory also. The winning team always destroys more turrets and very often deals more damage. Only after these values we get more in-depth observations at depth 4: **GDM**, **DRA%**, **TD@15**, **K:D** as well as **Wards cleared per minute**.\n",
    "\n",
    "In the last depth we get to see the more niche classification steps. These are the ones making a decision between a close matchup. There we can see values such as **DRA@15**, **Kills / game**, **HER%**, and **FB%**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy calculated from the test set = 76.69 %\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "Blue team win       0.74      0.81      0.77       144\n",
      " Red team win       0.80      0.73      0.76       152\n",
      "\n",
      "     accuracy                           0.77       296\n",
      "    macro avg       0.77      0.77      0.77       296\n",
      " weighted avg       0.77      0.77      0.77       296\n",
      "\n"
     ]
    }
   ],
   "source": [
    "predictions = classifier.predict(test_data)\n",
    "\n",
    "accuracy = accuracy_score(test_labels, predictions)\n",
    "print (f\"Accuracy calculated from the test set = {round(accuracy * 100, 2)} %\")\n",
    "print (classification_report(test_labels, predictions, target_names=['Blue team win', 'Red team win']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The decision tree also seemed to give good results and be accurate, but with like MultinomialNB I wasn't satisfied with the results it give when I tried predicting with probabilities. The results seemed to be out of touch with reality and seemed overfitting. The model can still be used to observe the variables it uses to for classification."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.4 Linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Variable</th>\n",
       "      <th>coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>K:D_y</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>HERPG_x</td>\n",
       "      <td>0.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>K:D_x</td>\n",
       "      <td>0.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>VWPM_y</td>\n",
       "      <td>0.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>VWPM_x</td>\n",
       "      <td>0.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>DRA@15_y</td>\n",
       "      <td>0.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>TD@15_y</td>\n",
       "      <td>0.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>Towers killed_y</td>\n",
       "      <td>0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>CSM_y</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DRAPG_x</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Variable  coefficient\n",
       "42            K:D_y         0.28\n",
       "14          HERPG_x         0.25\n",
       "15            K:D_x         0.16\n",
       "50           VWPM_y         0.13\n",
       "23           VWPM_x         0.10\n",
       "30         DRA@15_y         0.10\n",
       "47          TD@15_y         0.09\n",
       "48  Towers killed_y         0.06\n",
       "27            CSM_y         0.05\n",
       "4           DRAPG_x         0.05"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler = StandardScaler()\n",
    "reg = linear_model.LinearRegression()\n",
    "reg.fit(train_data, train_labels)\n",
    "\n",
    "mostUseful = pd.DataFrame()\n",
    "mostUseful[\"Variable\"] = colnames\n",
    "mostUseful[\"coefficient\"] = [round(coef, 2) for coef in reg.coef_]\n",
    "mostUseful = mostUseful.sort_values('coefficient', ascending = False)\n",
    "mostUseful.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Variable</th>\n",
       "      <th>coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>NASHPG_x</td>\n",
       "      <td>-0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>HERPG_y</td>\n",
       "      <td>-0.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>WCPM_y</td>\n",
       "      <td>-0.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>DRAPG_y</td>\n",
       "      <td>-0.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>TD@15_x</td>\n",
       "      <td>-0.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Kills / game_x</td>\n",
       "      <td>-0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>WPM_y</td>\n",
       "      <td>-0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>WPM_x</td>\n",
       "      <td>-0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>Towers lost_y</td>\n",
       "      <td>-0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DRA@15_x</td>\n",
       "      <td>-0.03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Variable  coefficient\n",
       "18        NASHPG_x        -0.36\n",
       "41         HERPG_y        -0.24\n",
       "51          WCPM_y        -0.22\n",
       "31         DRAPG_y        -0.20\n",
       "20         TD@15_x        -0.11\n",
       "16  Kills / game_x        -0.05\n",
       "52           WPM_y        -0.04\n",
       "25           WPM_x        -0.04\n",
       "49   Towers lost_y        -0.03\n",
       "3         DRA@15_x        -0.03"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mostUseful = mostUseful.sort_values('coefficient', ascending = True)\n",
    "mostUseful.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Linear regression gives us somewhat bizarre results. This model says that statistics like red team's **K:D** and **DRA@15** would positively indicate the blue team's win. This of course makes no sense in reality. Let's take a look at the model's metrics:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE on test set =  0.20621119663978393\n",
      "R2s on test set =  0.1745522567255431\n",
      "RMSE on test set =  0.4541048300115117\n",
      "RMSE on train set =  0.41163324191309236\n"
     ]
    }
   ],
   "source": [
    "pred_test_labels = reg.predict(test_data)\n",
    "pred_labels = reg.predict(train_data)\n",
    "\n",
    "mse_test = mean_squared_error(test_labels, pred_test_labels)\n",
    "r2s_test = r2_score(test_labels, pred_test_labels) \n",
    "print(\"MSE on test set = \", mse_test)\n",
    "print(\"R2s on test set = \", r2s_test)\n",
    "rmse_test = np.sqrt(mse_test)\n",
    "print(\"RMSE on test set = \", rmse_test)\n",
    "\n",
    "mse_train = mean_squared_error(train_labels, pred_labels)\n",
    "rmse_train = np.sqrt(mse_train)\n",
    "print(\"RMSE on train set = \", rmse_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the R-squared value is extremely low. It should be around 0.7 to indicate good fitting of a linear model. So these results cannot be taken as valid points."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.5 TensorFlow Keras Sequential\n",
    "\n",
    "With the neural network I decided to go for a simple model. It has a normalization layer and two hidden layers with 24 and 12, as well as one <code>Dropout</code> layer between them with a rate of 0.9 to prevent overfitting. For the <code>BATCH_SIZE</code> I chose 64, which seemed to give me the best results with my data set of 1477 games. I run 500 <code>EPOCHS</code>, which takes only around 35 seconds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_7\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " normalization_7 (Normalizat  (None, 54)               109       \n",
      " ion)                                                            \n",
      "                                                                 \n",
      " dense_14 (Dense)            (None, 24)                1320      \n",
      "                                                                 \n",
      " dropout_9 (Dropout)         (None, 24)                0         \n",
      "                                                                 \n",
      " dense_15 (Dense)            (None, 12)                300       \n",
      "                                                                 \n",
      " output (Dense)              (None, 1)                 13        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,742\n",
      "Trainable params: 1,633\n",
      "Non-trainable params: 109\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "normalizer = layers.Normalization()\n",
    "normalizer.adapt(np.array(data))\n",
    "\n",
    "def makeModel():    \n",
    "    model = models.Sequential([\n",
    "        normalizer,\n",
    "        layers.Dense(24, activation = \"relu\", input_shape = (54,)),\n",
    "        layers.Dropout(0.9),\n",
    "        layers.Dense(12, activation = \"relu\"),\n",
    "        layers.Dense(1, activation = \"sigmoid\", name = \"output\")\n",
    "    ])\n",
    "    \n",
    "    model.compile(\n",
    "        optimizer = optimizers.RMSprop(),\n",
    "        loss = losses.BinaryCrossentropy(),\n",
    "        metrics = [\n",
    "            metrics.BinaryAccuracy(name = 'acc'),\n",
    "            metrics.Recall(name = 'recall'), \n",
    "        ]\n",
    "    )\n",
    "    \n",
    "    return model\n",
    "\n",
    "model = makeModel()\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total elapsed time 18.77 s and 0.038 s/epoch\n"
     ]
    }
   ],
   "source": [
    "EPOCHS = 500\n",
    "BATCH_SIZE = 64\n",
    "\n",
    "t0 = time.time()\n",
    "history = model.fit(\n",
    "    train_data,\n",
    "    train_labels,\n",
    "    epochs = EPOCHS,\n",
    "    batch_size = BATCH_SIZE,\n",
    "    validation_data = (validation_data, validation_labels),\n",
    "    verbose = 0\n",
    ")\n",
    "t1 = time.time()\n",
    "etime = t1 - t0\n",
    "print(f'Total elapsed time {etime:.2f} s and {etime / EPOCHS:.3f} s/epoch')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd1f49ae-1022-4f89-8883-24e6d841c3c3",
   "metadata": {},
   "source": [
    "## 5. Validation and overfitting\n",
    "\n",
    "To see how well the model performed and to ensure that no overfitting is happening we check the validation metrics. Here we see that the validation results are quite close to the training results, this is good since too much difference between these results means that something may have gone wrong."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 74.00 % \t\t Val Accuracy: 71.00 %\n",
      "Recall: 86.00 % \t\t Val Recall: 87.00 %\n"
     ]
    }
   ],
   "source": [
    "print(f\"Accuracy: {round(history.history['acc'][-1], 2) * 100:.2f} % \\t\\t Val Accuracy: {round(history.history['val_acc'][-1], 2) * 100:.2f} %\")\n",
    "print(f\"Recall: {round(history.history['recall'][-1], 2) * 100:.2f} % \\t\\t Val Recall: {round(history.history['val_recall'][-1], 2) * 100:.2f} %\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_loss(history):\n",
    "    plt.plot(history.epoch, history.history['loss'])\n",
    "    plt.plot(history.epoch, history.history['val_loss'], linestyle=\"--\")\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.grid()\n",
    "\n",
    "plot_loss(history)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the loss plot we can also see that the validation curve stays under the training curve and has a similar curvature to the training curve. If the validation curve would've gone above the training curve, it would be an indication of possible overfitting. This model seems not to be overfitting. The chance of overfitting is lowered a lot by the dropout layer in the model and the normalization layer."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c042a705-6922-4ad9-af15-5832b993e76c",
   "metadata": {},
   "source": [
    "## 6. Performance and evaluation\n",
    "\n",
    "In this section we are taking a look at the loss, accuracy and recall values on each epoch with plots for both training and validation.\n",
    "\n",
    "### 6.1 Evaluating plots in training and validation\n",
    "\n",
    "Validating the model outputs are important to ensure the model's reliability. The main aim of validating the model in the last section is that it provides an opportunity for us to improve the settings. In this section we look at all the metrics more in-depth, since in the training section we looked only at the loss plot to determine is the model overfitting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1440x720 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs = history.epoch\n",
    "loss = history.history['loss']\n",
    "acc = history.history['acc']\n",
    "recall = history.history['recall']\n",
    "val_loss = history.history['val_loss']\n",
    "val_acc = history.history['val_acc']\n",
    "val_recall = history.history['val_recall']\n",
    "\n",
    "# Plot all of the metrics\n",
    "fig, axs = plt.subplots(2, 3)\n",
    "fig.subplots_adjust(hspace=0.3, wspace=0.2)\n",
    "axs[0, 0].plot(epochs, loss)\n",
    "axs[0, 0].set_title('Train Loss')\n",
    "axs[0, 0].grid()\n",
    "axs[0, 1].plot(epochs, acc, 'tab:orange')\n",
    "axs[0, 1].set_title('Train Accuracy')\n",
    "axs[0, 1].grid()\n",
    "axs[0, 2].plot(epochs, recall, 'tab:purple')\n",
    "axs[0, 2].set_title('Train recall')\n",
    "axs[0, 2].grid()\n",
    "axs[1, 0].plot(epochs, val_loss, 'tab:green')\n",
    "axs[1, 0].set_title('Validation loss')\n",
    "axs[1, 0].grid()\n",
    "axs[1, 1].plot(epochs, val_acc, 'tab:red')\n",
    "axs[1, 1].set_title('Validation accuracy')\n",
    "axs[1, 1].grid()\n",
    "axs[1, 2].plot(epochs, val_recall, 'tab:pink')\n",
    "axs[1, 2].set_title('Validation recall')\n",
    "axs[1, 2].grid()\n",
    "\n",
    "for ax in axs.flat:\n",
    "    ax.set(xlabel='Epoch', ylabel='Value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "ba84bca6-365d-4aca-bc23-975dc9c9464f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss = 0.581\n",
      "Test accuracy = 0.696\n",
      "Test recall = 0.789\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the trained model with test dataset\n",
    "test_loss, test_accuracy, test_recall = model.evaluate(test_data, test_labels, verbose = 0)\n",
    "print(f'Test loss = {test_loss:.3f}')\n",
    "print(f'Test accuracy = {test_accuracy:.3f}')\n",
    "print(f'Test recall = {test_recall:.3f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7. Conclusion\n",
    "\n",
    "With the <code>Keras Sequential</code> model we get a respectable 70% accuracy and recall of 80% on the test set. We also achieve a pretty small loss value and avoid overfitting. I was also happy with the results of the custom predictions I ran with it, as it clearly favors the stronger teams as it should. This model will be the one we will use in the demo section."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8. Predict games demo\n",
    "\n",
    "This is the demo for the prediction system. It takes each team's current strenght from [gol.gg](https://gol.gg/esports/home/) and predicts the probablities of a win with the <code>Keras Sequential</code> model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: predictor\\assets\n"
     ]
    }
   ],
   "source": [
    "model.save(\"predictor\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_sec(time_str):\n",
    "    m, s = time_str.split(':')\n",
    "    return float(60 * int(m) + int(s))\n",
    "\n",
    "def predict(team1, team2, region):\n",
    "    predictingModel = model\n",
    "    url=''\n",
    "    if region == \"LEC\":\n",
    "        url = 'https://gol.gg/teams/list/season-ALL/split-ALL/tournament-LEC%20Spring%202022/'\n",
    "    if region == \"LCS\":\n",
    "        url = 'https://gol.gg/teams/list/season-ALL/split-ALL/tournament-LCS%20Spring%202022/'\n",
    "    if region == \"LCK\":\n",
    "        url = 'https://gol.gg/teams/list/season-ALL/split-ALL/tournament-LCK%20Spring%202022/'\n",
    "\n",
    "    data = pd.read_html(requests.get(url, headers={'User-agent': 'your bot 0.1'}).text)[1]\n",
    "\n",
    "    data[\"FT%\"] = data[\"FT%\"].replace('-', float(0)).astype(float)\n",
    "    data[\"Win rate\"] = data[\"Win rate\"].apply(lambda x: float(x.replace(\"%\", \"\")))\n",
    "    data[\"Game duration\"] = data[\"Game duration\"].apply(lambda x: get_sec(x))\n",
    "    data[\"FB%\"] = data[\"FB%\"].replace('-', float(0)).astype(float)\n",
    "    data[\"Season\"] = data[\"Season\"].apply(lambda x: str(x).replace('S', '')).astype(float)\n",
    "    data = data.reindex(sorted(data.columns), axis=1)\n",
    "\n",
    "    teamOne = data[data[\"Name\"] == team1]\n",
    "    teamTwo = data[data[\"Name\"] == team2]\n",
    "\n",
    "    predictionsData = pd.merge(teamOne, teamTwo, left_on=\"Region\", right_on=\"Region\")\n",
    "    predictionsData = predictionsData.drop(columns=['Name_x', 'Name_y', 'Region'])\n",
    "    predictionsData = predictionsData.astype(float)\n",
    "\n",
    "    predictions = predictingModel.predict(predictionsData)\n",
    "    team1Result = predictions[0][0]\n",
    "    team2Result = (1 -predictions[0][0])\n",
    "\n",
    "    if (team1Result > team2Result):\n",
    "        print(f\"{team1Result * 100:.2f}% {team1} - {team2Result * 100:.2f}% {team2}\\n\")\n",
    "    else:\n",
    "        print(f\"{team2Result * 100:.2f}% {team2} - {team1Result * 100:.2f}% {team1}\\n\")\n",
    "    data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "76.20% G2 Esports - 23.80% Astralis\n",
      "\n",
      "69.12% Excel Esports - 30.88% Team BDS\n",
      "\n",
      "62.79% MAD Lions - 37.21% SK Gaming\n",
      "\n",
      "55.03% Rogue - 44.97% Misfits Gaming\n",
      "\n",
      "65.73% G2 Esports - 34.27% Astralis\n",
      "\n",
      "58.60% Fnatic - 41.40% Team Vitality\n",
      "\n"
     ]
    }
   ],
   "source": [
    "predict(\"Astralis\", \"G2 Esports\", \"LEC\")\n",
    "predict(\"Excel Esports\", \"Team BDS\", \"LEC\")\n",
    "predict(\"MAD Lions\", \"SK Gaming\", \"LEC\")\n",
    "predict(\"Rogue\", \"Misfits Gaming\", \"LEC\")\n",
    "predict(\"G2 Esports\", \"Astralis\", \"LEC\")\n",
    "predict(\"Team Vitality\", \"Fnatic\", \"LEC\")"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "ee60405de99a25455e5b9a69ae653e73d758a5fb2b9c4923700ed5732cb6ef82"
  },
  "kernelspec": {
   "display_name": "Python 3.9.7 64-bit (conda)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}