FPL Transfer Optimizer

Get mathematically optimal transfer suggestions for your Fantasy Premier League team.

What It Does

Enter your FPL team ID → Get optimal transfer recommendations based on:

• Expected points over the next 3 gameweeks
• Budget constraints (including selling prices)
• Squad composition rules (positions, team limits)
• Point hit analysis (-4 penalty for extra transfers)

Uses Mixed Integer Linear Programming (MILP) to guarantee mathematically optimal results.

Tech Stack

• Frontend: Next.js 15 with TypeScript, Tailwind CSS
• Optimization: Python with PuLP and HiGHS MILP solver
• Hosting: Vercel

Quick Start

Prerequisites

• Node.js 18+
• Python 3.9+

Installation

1. Clone the repository

   git clone https://github.com/Shirazik/fpl-optimizer.git
   cd fpl-optimizer

2. Install dependencies

   npm install

3. Set up Python environment (for optimizer)

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   pip install -r python/requirements.txt

4. Run the development server

   npm run dev

5. Open your browser Navigate to http://localhost:3000

How to Use

1. Find your FPL team ID:

   • Go to Fantasy Premier League
   • Log in and go to your team page
   • Look at the URL: fantasy.premierleague.com/entry/YOUR-TEAM-ID/event/XX
   • Copy the number after /entry/

2. Enter your team ID on the home page

3. Click "Get Transfer Suggestions"

4. View your optimal transfers with:

   • Players to transfer in/out
   • Expected points gain
   • Whether a point hit is worth it

How It Works

MILP Optimization

The optimizer uses Mixed Integer Linear Programming to find mathematically optimal transfers.

Decision Variables:

• x[player] = 1 if player is in final squad, 0 otherwise
• t_in[player] = 1 if player transferred in
• t_out[player] = 1 if player transferred out

Objective Function:

Maximize: Σ(expected_points × horizon_weight) - point_hit_penalty

Constraints:

• Squad size = 15 players
• Positions: 2 GK, 5 DEF, 5 MID, 3 FWD
• Budget: Total cost ≤ available budget (including selling prices)
• Team limit: Max 3 players per Premier League team
• Transfer balance: Players in = Players out

Horizon Weights: Future gameweeks are weighted less due to prediction uncertainty:

GW1: 1.0, GW2: 0.85, GW3: 0.7

Expected Points Calculation

Currently uses a form-based model:

base_ep = (form + points_per_game) / 2
adjusted_ep = base_ep × fixture_multiplier × availability_factor

Factors in:

• Recent form rating
• Points per game average
• Fixture difficulty
• Player availability (injured/suspended)

Selling Price Calculation

Implements FPL's "half-profit rule":

• If price rises after purchase: selling price = purchase price + (profit / 2)
• If price falls: selling price = current price
• Example: Buy at £10.0m, rises to £10.4m → Sell at £10.2m

Project Structure

fpl-optimizer/
├── app/
│   ├── page.tsx                      # Home page (team ID input)
│   ├── optimize/[teamId]/page.tsx    # Optimization results page
│   ├── layout.tsx                    # Root layout
│   ├── globals.css                   # Styles
│   └── api/
│       ├── optimize/transfer/        # Optimization endpoint
│       └── fpl/                      # FPL data endpoints
├── components/
│   └── transfers/                    # Transfer suggestion components
├── lib/
│   ├── fpl-api.ts                   # FPL API client
│   └── optimizer.ts                 # Python script caller
├── python/
│   ├── optimize_transfers.py        # MILP optimizer
│   └── requirements.txt             # Python dependencies
└── types/                           # TypeScript type definitions

Development

Available Commands

npm run dev      # Start development server
npm run build    # Build for production
npm run start    # Start production server
npm run lint     # Run ESLint

Testing the Optimizer

# Activate Python environment
source venv/bin/activate

# Test optimizer directly
python python/optimize_transfers.py < test_input.json

API Endpoints

Optimize Transfers

POST /api/optimize/transfer
Content-Type: application/json

{
  "teamId": "123456",
  "maxTransfers": 2,
  "horizon": 3
}

Returns optimal transfer suggestions with expected point gains.

Deployment

Deploy to Vercel

1. Push code to GitHub
2. Import repository in Vercel
3. Deploy

Vercel will auto-detect Next.js configuration.

Troubleshooting

"Team not found" error:

• Double-check your team ID
• Make sure you're using the number from the URL

"This team is private" error:

• The team owner has set their team to private
• Only public teams can be analyzed

Optimization fails:

• Make sure Python 3.9+ is installed
• Verify virtual environment is activated: source venv/bin/activate
• Check dependencies are installed: pip install -r python/requirements.txt

Why MILP?

FPL transfer optimization is a constrained optimization problem. MILP solvers:

• Guarantee mathematical optimality
• Handle complex constraints efficiently
• Scale better than brute force or genetic algorithms
• Provide deterministic, reproducible results

License

MIT License - see LICENSE

Acknowledgments

• Fantasy Premier League for the official API
• PuLP and HiGHS teams for optimization tools
• The FPL community for inspiration

---

Built for FPL managers who want to win ⚽