viz.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import inference\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_path = \"/data/nfl/armchairanalysis/crewpool/joined.parquet\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>gid</th>\n",
       "      <th>off</th>\n",
       "      <th>def</th>\n",
       "      <th>ptso</th>\n",
       "      <th>ptsd</th>\n",
       "      <th>time_elapsed_sec</th>\n",
       "      <th>seas</th>\n",
       "      <th>sprv</th>\n",
       "      <th>ou</th>\n",
       "      <th>home_dog</th>\n",
       "      <th>favorite</th>\n",
       "      <th>underdog</th>\n",
       "      <th>final_points_favorite</th>\n",
       "      <th>final_points_underdog</th>\n",
       "      <th>did_favorite_cover</th>\n",
       "      <th>current_points_favorite</th>\n",
       "      <th>current_points_underdog</th>\n",
       "      <th>current_points_underdog_with_spread</th>\n",
       "      <th>current_spread_difference</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>84060</td>\n",
       "      <td>519</td>\n",
       "      <td>NYG</td>\n",
       "      <td>SF</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2002</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>38.5</td>\n",
       "      <td>1</td>\n",
       "      <td>SF</td>\n",
       "      <td>NYG</td>\n",
       "      <td>16</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84061</td>\n",
       "      <td>519</td>\n",
       "      <td>NYG</td>\n",
       "      <td>SF</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2002</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>38.5</td>\n",
       "      <td>1</td>\n",
       "      <td>SF</td>\n",
       "      <td>NYG</td>\n",
       "      <td>16</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84062</td>\n",
       "      <td>519</td>\n",
       "      <td>NYG</td>\n",
       "      <td>SF</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>2002</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>38.5</td>\n",
       "      <td>1</td>\n",
       "      <td>SF</td>\n",
       "      <td>NYG</td>\n",
       "      <td>16</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84063</td>\n",
       "      <td>519</td>\n",
       "      <td>NYG</td>\n",
       "      <td>SF</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>2002</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>38.5</td>\n",
       "      <td>1</td>\n",
       "      <td>SF</td>\n",
       "      <td>NYG</td>\n",
       "      <td>16</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84064</td>\n",
       "      <td>519</td>\n",
       "      <td>NYG</td>\n",
       "      <td>SF</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>2002</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>38.5</td>\n",
       "      <td>1</td>\n",
       "      <td>SF</td>\n",
       "      <td>NYG</td>\n",
       "      <td>16</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gid  off def  ptso  ptsd  time_elapsed_sec  seas  sprv    ou  home_dog  \\\n",
       "84060  519  NYG  SF     0     0                 0  2002  -3.5  38.5         1   \n",
       "84061  519  NYG  SF     0     0                 0  2002  -3.5  38.5         1   \n",
       "84062  519  NYG  SF     0     0                35  2002  -3.5  38.5         1   \n",
       "84063  519  NYG  SF     0     0                40  2002  -3.5  38.5         1   \n",
       "84064  519  NYG  SF     0     0                40  2002  -3.5  38.5         1   \n",
       "\n",
       "      favorite underdog  final_points_favorite  final_points_underdog  \\\n",
       "84060       SF      NYG                     16                     13   \n",
       "84061       SF      NYG                     16                     13   \n",
       "84062       SF      NYG                     16                     13   \n",
       "84063       SF      NYG                     16                     13   \n",
       "84064       SF      NYG                     16                     13   \n",
       "\n",
       "       did_favorite_cover  current_points_favorite  current_points_underdog  \\\n",
       "84060                   0                        0                        0   \n",
       "84061                   0                        0                        0   \n",
       "84062                   0                        0                        0   \n",
       "84063                   0                        0                        0   \n",
       "84064                   0                        0                        0   \n",
       "\n",
       "       current_points_underdog_with_spread  current_spread_difference  \n",
       "84060                                  3.5                       -3.5  \n",
       "84061                                  3.5                       -3.5  \n",
       "84062                                  3.5                       -3.5  \n",
       "84063                                  3.5                       -3.5  \n",
       "84064                                  3.5                       -3.5  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_parquet(input_path)\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>time_elapsed_sec</th>\n",
       "      <th>sprv</th>\n",
       "      <th>home_dog</th>\n",
       "      <th>current_points_favorite</th>\n",
       "      <th>current_points_underdog</th>\n",
       "      <th>current_spread_difference</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>84060</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84061</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84062</td>\n",
       "      <td>35</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84063</td>\n",
       "      <td>40</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84064</td>\n",
       "      <td>40</td>\n",
       "      <td>-3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       time_elapsed_sec  sprv  home_dog  current_points_favorite  \\\n",
       "84060                 0  -3.5         1                        0   \n",
       "84061                 0  -3.5         1                        0   \n",
       "84062                35  -3.5         1                        0   \n",
       "84063                40  -3.5         1                        0   \n",
       "84064                40  -3.5         1                        0   \n",
       "\n",
       "       current_points_underdog  current_spread_difference  \n",
       "84060                        0                       -3.5  \n",
       "84061                        0                       -3.5  \n",
       "84062                        0                       -3.5  \n",
       "84063                        0                       -3.5  \n",
       "84064                        0                       -3.5  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for_inference = df.drop(\n",
    "    ['gid', 'seas', 'off', 'def', 'favorite', 'underdog', 'final_points_favorite', \n",
    "     'final_points_underdog', 'ptso', 'ptsd', 'ou', \"did_favorite_cover\",\n",
    "     'current_points_underdog_with_spread'\n",
    "    ]\n",
    "    , axis=1)\n",
    "for_inference.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "MODEL_PATH = 'model/model.h5'\n",
    "SCALER_PATH = 'model/scaler.pkl'\n",
    "\n",
    "model = inference.load_model(MODEL_PATH)\n",
    "scaler = inference.load_scaler(SCALER_PATH)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(568306, 6)\n"
     ]
    },
    {
     "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>-1.762735</td>\n",
       "      <td>-1.038608</td>\n",
       "      <td>1.461786</td>\n",
       "      <td>-1.16769</td>\n",
       "      <td>-1.049051</td>\n",
       "      <td>-0.103136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>-1.762735</td>\n",
       "      <td>-1.038608</td>\n",
       "      <td>1.461786</td>\n",
       "      <td>-1.16769</td>\n",
       "      <td>-1.049051</td>\n",
       "      <td>-0.103136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>-1.729607</td>\n",
       "      <td>-1.038608</td>\n",
       "      <td>1.461786</td>\n",
       "      <td>-1.16769</td>\n",
       "      <td>-1.049051</td>\n",
       "      <td>-0.103136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>-1.724874</td>\n",
       "      <td>-1.038608</td>\n",
       "      <td>1.461786</td>\n",
       "      <td>-1.16769</td>\n",
       "      <td>-1.049051</td>\n",
       "      <td>-0.103136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>-1.724874</td>\n",
       "      <td>-1.038608</td>\n",
       "      <td>1.461786</td>\n",
       "      <td>-1.16769</td>\n",
       "      <td>-1.049051</td>\n",
       "      <td>-0.103136</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          0         1         2        3         4         5\n",
       "0 -1.762735 -1.038608  1.461786 -1.16769 -1.049051 -0.103136\n",
       "1 -1.762735 -1.038608  1.461786 -1.16769 -1.049051 -0.103136\n",
       "2 -1.729607 -1.038608  1.461786 -1.16769 -1.049051 -0.103136\n",
       "3 -1.724874 -1.038608  1.461786 -1.16769 -1.049051 -0.103136\n",
       "4 -1.724874 -1.038608  1.461786 -1.16769 -1.049051 -0.103136"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled = scaler.transform(for_inference)\n",
    "print(scaled.shape)\n",
    "pd.DataFrame(scaled).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(568306, 1)\n"
     ]
    },
    {
     "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>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.484616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.484616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.484979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.485030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.485030</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          0\n",
       "0  0.484616\n",
       "1  0.484616\n",
       "2  0.484979\n",
       "3  0.485030\n",
       "4  0.485030"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = model.predict(scaled)\n",
    "print(predictions.shape)\n",
    "pd.DataFrame(predictions).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>gid</th>\n",
       "      <th>off</th>\n",
       "      <th>def</th>\n",
       "      <th>ptso</th>\n",
       "      <th>ptsd</th>\n",
       "      <th>time_elapsed_sec</th>\n",
       "      <th>seas</th>\n",
       "      <th>sprv</th>\n",
       "      <th>ou</th>\n",
       "      <th>home_dog</th>\n",
       "      <th>...</th>\n",
       "      <th>final_points_underdog</th>\n",
       "      <th>did_favorite_cover</th>\n",
       "      <th>current_points_favorite</th>\n",
       "      <th>current_points_underdog</th>\n",
       "      <th>current_points_underdog_with_spread</th>\n",
       "      <th>current_spread_difference</th>\n",
       "      <th>cover_prob</th>\n",
       "      <th>prediction</th>\n",
       "      <th>time_elapsed_min</th>\n",
       "      <th>predicted_correctly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>240115</td>\n",
       "      <td>1473</td>\n",
       "      <td>BUF</td>\n",
       "      <td>SD</td>\n",
       "      <td>3</td>\n",
       "      <td>28</td>\n",
       "      <td>1710</td>\n",
       "      <td>2005</td>\n",
       "      <td>10.5</td>\n",
       "      <td>42.5</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>3</td>\n",
       "      <td>13.5</td>\n",
       "      <td>14.5</td>\n",
       "      <td>0.947201</td>\n",
       "      <td>True</td>\n",
       "      <td>28.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>474870</td>\n",
       "      <td>2924</td>\n",
       "      <td>IND</td>\n",
       "      <td>HOU</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>2909</td>\n",
       "      <td>2011</td>\n",
       "      <td>8.5</td>\n",
       "      <td>43.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "      <td>8.5</td>\n",
       "      <td>25.5</td>\n",
       "      <td>0.999997</td>\n",
       "      <td>True</td>\n",
       "      <td>48.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>382250</td>\n",
       "      <td>2355</td>\n",
       "      <td>OAK</td>\n",
       "      <td>HOU</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>38</td>\n",
       "      <td>2008</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.457300</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>159829</td>\n",
       "      <td>979</td>\n",
       "      <td>BUF</td>\n",
       "      <td>NYJ</td>\n",
       "      <td>17</td>\n",
       "      <td>6</td>\n",
       "      <td>3163</td>\n",
       "      <td>2003</td>\n",
       "      <td>3.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>6</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.910462</td>\n",
       "      <td>True</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>179955</td>\n",
       "      <td>1104</td>\n",
       "      <td>GB</td>\n",
       "      <td>NYG</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>2246</td>\n",
       "      <td>2004</td>\n",
       "      <td>7.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>14.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.197314</td>\n",
       "      <td>False</td>\n",
       "      <td>37.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>490733</td>\n",
       "      <td>3020</td>\n",
       "      <td>KC</td>\n",
       "      <td>OAK</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>2349</td>\n",
       "      <td>2011</td>\n",
       "      <td>4.0</td>\n",
       "      <td>40.5</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>21</td>\n",
       "      <td>25.0</td>\n",
       "      <td>-25.0</td>\n",
       "      <td>0.001777</td>\n",
       "      <td>False</td>\n",
       "      <td>39.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>460589</td>\n",
       "      <td>2836</td>\n",
       "      <td>DEN</td>\n",
       "      <td>KC</td>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>2531</td>\n",
       "      <td>2010</td>\n",
       "      <td>9.0</td>\n",
       "      <td>48.5</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>12.0</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>0.525593</td>\n",
       "      <td>True</td>\n",
       "      <td>42.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>514210</td>\n",
       "      <td>3162</td>\n",
       "      <td>WAS</td>\n",
       "      <td>PHI</td>\n",
       "      <td>10</td>\n",
       "      <td>34</td>\n",
       "      <td>3571</td>\n",
       "      <td>2011</td>\n",
       "      <td>8.0</td>\n",
       "      <td>45.5</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>10</td>\n",
       "      <td>18.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>0.999924</td>\n",
       "      <td>True</td>\n",
       "      <td>59.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>523316</td>\n",
       "      <td>3217</td>\n",
       "      <td>MIA</td>\n",
       "      <td>OAK</td>\n",
       "      <td>14</td>\n",
       "      <td>10</td>\n",
       "      <td>2420</td>\n",
       "      <td>2012</td>\n",
       "      <td>-2.5</td>\n",
       "      <td>39.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>16.5</td>\n",
       "      <td>-6.5</td>\n",
       "      <td>0.279497</td>\n",
       "      <td>False</td>\n",
       "      <td>40.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>260957</td>\n",
       "      <td>1601</td>\n",
       "      <td>MIN</td>\n",
       "      <td>WAS</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>1007</td>\n",
       "      <td>2006</td>\n",
       "      <td>4.5</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>10.5</td>\n",
       "      <td>-7.5</td>\n",
       "      <td>0.360444</td>\n",
       "      <td>False</td>\n",
       "      <td>16.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         gid  off  def  ptso  ptsd  time_elapsed_sec  seas  sprv    ou  \\\n",
       "240115  1473  BUF   SD     3    28              1710  2005  10.5  42.5   \n",
       "474870  2924  IND  HOU     0    34              2909  2011   8.5  43.0   \n",
       "382250  2355  OAK  HOU     0     0                38  2008  -7.0  43.0   \n",
       "159829   979  BUF  NYJ    17     6              3163  2003   3.0  36.0   \n",
       "179955  1104   GB  NYG     7     7              2246  2004   7.0  44.0   \n",
       "490733  3020   KC  OAK    21     0              2349  2011   4.0  40.5   \n",
       "460589  2836  DEN   KC     3    10              2531  2010   9.0  48.5   \n",
       "514210  3162  WAS  PHI    10    34              3571  2011   8.0  45.5   \n",
       "523316  3217  MIA  OAK    14    10              2420  2012  -2.5  39.0   \n",
       "260957  1601  MIN  WAS     6     3              1007  2006   4.5  35.0   \n",
       "\n",
       "        home_dog  ... final_points_underdog did_favorite_cover  \\\n",
       "240115         0  ...                    10                  1   \n",
       "474870         0  ...                     7                  1   \n",
       "382250         1  ...                    27                  0   \n",
       "159829         0  ...                     6                  1   \n",
       "179955         0  ...                    14                  0   \n",
       "490733         0  ...                    28                  0   \n",
       "460589         0  ...                     6                  0   \n",
       "514210         0  ...                    10                  1   \n",
       "523316         1  ...                    35                  0   \n",
       "260957         0  ...                    19                  0   \n",
       "\n",
       "        current_points_favorite  current_points_underdog  \\\n",
       "240115                       28                        3   \n",
       "474870                       34                        0   \n",
       "382250                        0                        0   \n",
       "159829                       17                        6   \n",
       "179955                        7                        7   \n",
       "490733                        0                       21   \n",
       "460589                       10                        3   \n",
       "514210                       34                       10   \n",
       "523316                       10                       14   \n",
       "260957                        3                        6   \n",
       "\n",
       "        current_points_underdog_with_spread  current_spread_difference  \\\n",
       "240115                                 13.5                       14.5   \n",
       "474870                                  8.5                       25.5   \n",
       "382250                                  7.0                       -7.0   \n",
       "159829                                  9.0                        8.0   \n",
       "179955                                 14.0                       -7.0   \n",
       "490733                                 25.0                      -25.0   \n",
       "460589                                 12.0                       -2.0   \n",
       "514210                                 18.0                       16.0   \n",
       "523316                                 16.5                       -6.5   \n",
       "260957                                 10.5                       -7.5   \n",
       "\n",
       "        cover_prob  prediction  time_elapsed_min  predicted_correctly  \n",
       "240115    0.947201        True              28.0                    1  \n",
       "474870    0.999997        True              48.0                    1  \n",
       "382250    0.457300       False               0.0                    1  \n",
       "159829    0.910462        True              52.0                    1  \n",
       "179955    0.197314       False              37.0                    1  \n",
       "490733    0.001777       False              39.0                    1  \n",
       "460589    0.525593        True              42.0                    0  \n",
       "514210    0.999924        True              59.0                    1  \n",
       "523316    0.279497       False              40.0                    1  \n",
       "260957    0.360444       False              16.0                    1  \n",
       "\n",
       "[10 rows x 23 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['cover_prob'] = predictions\n",
    "df['prediction'] = predictions > 0.5\n",
    "df['time_elapsed_min'] = np.floor(df['time_elapsed_sec'] / 60)\n",
    "df['predicted_correctly'] = (df[\"did_favorite_cover\"]  == df.prediction.astype(int)).astype(int)\n",
    "df.sample(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>total_correct</th>\n",
       "      <th>total_games</th>\n",
       "      <th>prediction_accuracy</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time_elapsed_min</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0.0</td>\n",
       "      <td>6578</td>\n",
       "      <td>12467</td>\n",
       "      <td>0.527633</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>4212</td>\n",
       "      <td>7821</td>\n",
       "      <td>0.538550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2.0</td>\n",
       "      <td>4393</td>\n",
       "      <td>8099</td>\n",
       "      <td>0.542413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3.0</td>\n",
       "      <td>4656</td>\n",
       "      <td>8295</td>\n",
       "      <td>0.561302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4.0</td>\n",
       "      <td>4903</td>\n",
       "      <td>8486</td>\n",
       "      <td>0.577775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5.0</td>\n",
       "      <td>5064</td>\n",
       "      <td>8634</td>\n",
       "      <td>0.586518</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6.0</td>\n",
       "      <td>5045</td>\n",
       "      <td>8450</td>\n",
       "      <td>0.597041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7.0</td>\n",
       "      <td>5118</td>\n",
       "      <td>8440</td>\n",
       "      <td>0.606398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8.0</td>\n",
       "      <td>5183</td>\n",
       "      <td>8402</td>\n",
       "      <td>0.616877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9.0</td>\n",
       "      <td>5344</td>\n",
       "      <td>8487</td>\n",
       "      <td>0.629669</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  total_correct  total_games  prediction_accuracy\n",
       "time_elapsed_min                                                 \n",
       "0.0                        6578        12467             0.527633\n",
       "1.0                        4212         7821             0.538550\n",
       "2.0                        4393         8099             0.542413\n",
       "3.0                        4656         8295             0.561302\n",
       "4.0                        4903         8486             0.577775\n",
       "5.0                        5064         8634             0.586518\n",
       "6.0                        5045         8450             0.597041\n",
       "7.0                        5118         8440             0.606398\n",
       "8.0                        5183         8402             0.616877\n",
       "9.0                        5344         8487             0.629669"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grouped = df.groupby(\"time_elapsed_min\").agg(total_correct=('predicted_correctly', 'sum'), total_games=('predicted_correctly', 'count'),)\n",
    "grouped['prediction_accuracy'] = grouped['total_correct'] / grouped['total_games']\n",
    "grouped.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.figure(figsize=(15,110))\n",
    "to_plot = grouped[grouped.index < 60]  # Overtime has crazy results\n",
    "to_plot = to_plot[['prediction_accuracy']].rolling(4, min_periods=1).mean()  # Run moving average\n",
    "\n",
    "\n",
    "plt.figure(figsize=(15, 10))\n",
    "plt.rc('axes', labelsize=16) # Increase font for axis labels\n",
    "plt.plot(to_plot, color='darkolivegreen')\n",
    "plt.xlabel(\"Minutes Elapsed\")\n",
    "plt.ylabel(\"Model Accuracy\")\n",
    "\n",
    "# Add lines for each quarter\n",
    "plt.axvline(x=0, color='k',)\n",
    "plt.axvline(x=15, linestyle='', color='k',)\n",
    "plt.axvline(x=30, linestyle=':', color='k', alpha=0.50)\n",
    "plt.axvline(x=45, linestyle=':', color='k', alpha=0.50)\n",
    "plt.axvline(x=60, color='k',)\n",
    "\n",
    "# Background colors for sections\n",
    "background_color = \"green\"\n",
    "plt.axvspan(0, 15, alpha=0.05, color=background_color)\n",
    "plt.axvspan(15, 30, alpha=0.10, color=background_color, label='q1')\n",
    "plt.axvspan(30, 45, alpha=0.15, color=background_color)\n",
    "plt.axvspan(45, 60, alpha=0.20, color=background_color)\n",
    "\n",
    "plt.xticks(np.arange(0, 65, 5)) # Add ticks every 5 min\n",
    "\n",
    "\n",
    "plt.text(0.5, .85, '1st Quarter', rotation=90)\n",
    "plt.text(15.5, .85, '2nd Quarter', rotation=90)\n",
    "plt.text(30.5, .85, '3rd Quarter', rotation=90)\n",
    "plt.text(45.5, .85, '4th Quarter', rotation=90)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}