Dos caminos: construir el bot tú mismo o usar la plataforma en vivo para señales
Construye un bot de apuestas deportivas
desde cero
6 notebooks de Jupyter te llevan de cero a un bot de trading en vivo. Pensado para principiantes y diseñado para funcionar bien con Claude o ChatGPT.
Creado para Polymarket. Si no quieres construirlo, puedes usar la plataforma en vivo por separado con una cuenta gratis.
1.171 operaciones backtesteadas · 70.9% de acierto · Verificado on-chain · 7 deportes
Compra única. Descarga instantánea. Sin suscripción. ¿Prefieres señales en vivo? Empieza gratis con la plataforma.
La cuenta gratis incluye: panel en vivo + edge scanner, acceso retrasado a /v1/games y /v1/edges en 7 deportes, 500 llamadas API/mes. Sin tarjeta.
Works with AI assistants
New to Python? Every notebook is designed to work with Claude, ChatGPT, or any AI coding assistant. Paste any cell into an AI, ask "explain this" or "help me customize this for soccer," and get step-by-step guidance. You don't need to be a programmer to build a working bot.
Real backtest results from this system
| Sport | Trades | Win Rate | Avg Profit/Trade |
|---|---|---|---|
| NCAAMB | 821 | 73.1% | +13.8c |
| NBA | 203 | 66.5% | +8.5c |
| NHL | 147 | 64.6% | -1.2c |
| Combined | 1,171 | 70.9% | +11.0c |
Model v4 + Elo improvements (Mar 2026). Time-split validation against Polymarket bid/ask prices. Held to settlement, net of 2c fee + 1c slippage. Full statistical validation · Live results
What you'll build — module by module
6 Jupyter notebooks of working code. Here's a real cell from Module 3 — training a calibrated win-probability model on 60,000+ games.
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.calibration import CalibratedClassifierCV
# Train on 60,000+ games across 7 sports
model = GradientBoostingClassifier(n_estimators=200)
model.fit(X_train, y_train)
# Isotonic calibration for reliable probabilities
calibrated = CalibratedClassifierCV(model, method="isotonic")
calibrated.fit(X_cal, y_cal)
print(f"Brier: {brier_score(y_test, calibrated.predict_proba(X_test)[:, 1]):.4f}")Brier: 0.1395 | AUC: 0.890 | ECE: 0.033
Scraping ESPN
- Build an async data scraper for ESPN play-by-play
- Handle rate limits, retries, and API pagination
- Fetch 60,000+ games across 7 sports
- Store as efficient Parquet files
Production-grade data pipeline in 1 notebook
Try free ↓ · AI prompt: "Explain how this async scraper works"
Elo Ratings
- Implement Elo from scratch (no libraries)
- Home advantage, K-factor tuning, season resets
- 907+ NCAAMB teams, 258+ per sport
- Validate against known rankings
The simplest feature that matters most
AI prompt: "Help me add season-decay to my Elo system"
WP Models
- Train LR+Spline and XGBoost+Isotonic
- Isotonic calibration for probability accuracy
- Why simpler models beat complex ones for trading
- Sport-specific feature engineering
Worse Brier score = more trading profit
AI prompt: "Explain isotonic calibration like I'm a beginner"
Backtesting
- Time-split validation (train on N-1, test on N)
- Adverse selection and underdog traps
- Execution cost modeling (fees, slippage)
- Deduplication and subsampling
Every mistake that blows up your backtest
AI prompt: "Help me add a new sport to this backtest"
Live Bot
- Polymarket CLOB API integration
- ESPN adaptive polling (5s/15s)
- Edge detection and signal generation
- Order execution with shadow mode
A working bot you can run tonight
AI prompt: "Help me add Discord alerts to this bot"
Deployment
- FastAPI server with HTTPS
- Discord alert webhooks
- Cron scheduling and monitoring
- Cloudflare Tunnel for secure access
From laptop to 24/7 production
AI prompt: "Help me deploy this to a $7/mo VPS"
Want to see the actual code before buying?
Preview the first 8 cells of every module — real teaching, real code, not marketing copy.
Try Module 1 free
Scraping ESPN: build an async data pipeline that fetches 60,000+ games. Enter your email to download the full notebook instantly.
Check your email for the download link too.
No spam. Unsubscribe anytime.
Before and after
Before
- "I don't know Python but want to build a bot"
- Watching tutorials that stop at theory
- Buying picks from Discord that lose money
- No systematic edge, just gut feel
- No way to measure if a strategy works
After
- A working bot finding 8c+ edges 24/7
- AI-assisted coding — paste any cell and ask for help
- Data-driven decisions backed by 60k+ games
- Your own models, your own edge, your own system
- Full backtest framework proving it works before risking a dollar
Built by a trader, not a marketer
This course isn't theory — it's the exact system I built and run myself on Polymarket. The models, the backtesting framework, the deployment setup — I use all of it daily. The API at zenhodl.net runs the same code taught in these notebooks.
I built this because I couldn't find a prediction market course that showed real results with real code. Every "sports betting model" tutorial online stops at a confusion matrix. This one starts at the data pipeline and ends with a deployed bot executing real trades.
See live results at /results. Read the methodology at /methodology. Read our research paper.
Join 50+ traders in Discord
Course Q&A, live trading chat, strategy sharing, and edge signals. Join the server, then verify your Discord ID from your ZenHodl account to unlock paid roles.
Frequently asked questions
What do I need to get started? +
What skill level is this for? +
What you DO need: A computer (Mac, Windows, or Linux), an internet connection, and the willingness to follow instructions and experiment. The notebooks handle the rest.
Is there a refund policy? +
We're confident because the system works — 70.9% backtest win rate across 1,171 trades. Plus you can try Module 1 completely free before buying to make sure you like the teaching style.
Note: Refund requires demonstrated completion of all modules. Dataset purchases are non-refundable.
How do I use AI with this course? +
- Copy the cell into Claude, ChatGPT, or any AI assistant
- Ask: "Explain this code line by line" or "What does this function do?"
- To customize: "Help me modify this to track soccer instead of NBA"
- To debug: paste the error message and ask "How do I fix this?"
Think of it as having a patient tutor sitting next to you. The notebooks give you the working code — the AI helps you understand and extend it.
How long does it take to complete? +
Will this work for my sport? +
What's the difference between the course and the API? +
What can go wrong
Model degradation. Markets adapt. A model that works today may need retraining in 3-6 months as market efficiency improves.
Execution costs. Backtest results assume ~2-3c per trade in fees and slippage. Real execution can be worse in thin markets.
Small samples. Live trading started March 2026. The backtest is strong (1,171 trades) but live history is still growing.
Regime change. Rule changes, new market participants, or structural shifts can invalidate historical patterns.
Past performance does not guarantee future results. This is a tool for informed decision-making, not a guaranteed profit machine.
Start building your own edge
6 notebooks. Working code. Backtested: 1,171 trades at 70.9% WR.