Our Methodology

Why independent probabilities matter

Most sports analytics services derive "fair value" by devigging sportsbook lines — averaging Pinnacle, FanDuel, and DraftKings odds. This gives you a consensus probability that tracks the market by construction. It's useful for sports betting (finding +EV against soft books), but it cannot find prediction market mispricings — because the output already agrees with the market.

ZenHodl models are trained on game state only: score differential, seconds remaining, period, Elo ratings, and sport-specific features. No odds, no lines, no market prices are used as inputs. This makes our output genuinely independent from the market — when our fair probability diverges from the Polymarket ask price, that divergence is a real signal, not noise.

The tradeoff: our models can have a worse Brier score than market-derived models in absolute forecasting terms. But independence is what creates trading value when the output is validated against real market prices. See the validation page for the latest exported backtest snapshot and live results for current production performance.

Data pipeline

We scrape ESPN's play-by-play API across the core sports we model. Each game produces hundreds of snapshots — one per score change or significant event.

Sports: NBA, NCAAMB, NCAAWB, CFB, NFL, NHL, MLB

Data is stored as Apache Parquet files. One row = one game state (score, period, clock, ESPN WP, outcome label).

Module 1 of our course teaches you to build this exact scraper.

Feature engineering

Each game state is featurized with 13–16 variables. We deliberately keep the feature set small — overfitting to noise destroys trading value.

Model architecture

We use sport-specific models — no one-size-fits-all approach. The sections below distinguish the current live platform from adjacent research and course material.

Current live platform

The live platform currently covers NBA, NCAAMB, NCAAWB, CFB, NFL, NHL, MLB with sport-specific state models, calibration layers, and execution filters. We do not force one shared architecture across all sports.

Basketball & Football (NBA, NCAAMB, NCAAWB, CFB, NFL)

Logistic regression with isotonic spline calibration. We tested XGBoost, random forests, and neural nets. The key finding: simpler models produce better trading value because they don't overfit to noise in the training data. XGBoost achieves a better Brier score but worse c/trade. We select models by trading value, not accuracy.

Some sports justify extra complexity, but only when validation shows the added model risk is worth it. We optimize for durable trading value, not just a prettier forecast metric.

Hockey & Baseball (NHL, MLB)

NHL and MLB use state-aware probability models with sport-specific context features and calibration. They follow the same principle as the rest of the live platform: independent fair probabilities first, then market comparison and execution filtering.

Research and course extensions

Some additional model families appear in the course and research stack even when they are not part of the current 7-sport live product surface.

Soccer

Poisson goal models are a natural fit for three-way markets because they estimate home-win, draw, and away-win probabilities directly from scoring rates instead of forcing a binary approximation.

Esports (CS2, LoL)

Series markets in esports map cleanly to binomial or map-level models. These modules are part of the broader research/course ecosystem, but they are not the same thing as today’s live 7-sport product surface.

Spread/Total

Spread and total models require a different setup from moneyline markets: regression on remaining margin or total, then a distributional layer to convert that forecast into cover/over probabilities.

Backtesting methodology

Our backtests are designed to avoid the common mistakes that inflate results.

• Time-split validation: Train on season N−1, test on season N. No future data leaks.
• Real market prices where available: Where direct market snapshot history exists, backtests use recorded market prices. Where coverage is incomplete, we call out any proxy assumptions explicitly on the validation page.
• Deduplication: One entry per (game_id, side, score, period). No duplicate signals from the same game state.
• Subsampling where needed: Historical replay backtests can downsample state rows to better match how the live system would realistically observe and act on the game.
• Hold to settlement: All trades held to game end. No exit timing or stop-loss assumptions.
• Net of fees: Results are reported net of the fee and slippage assumptions used by the active backtest export, with the exact assumptions disclosed on the validation page.
• Sport-specific entry filters: Filters like score differential, game state, or pregame-consensus agreement are applied only where validation shows they improve robustness.
• Execution windows: Entry timing constraints vary by sport and market. We only keep them when out-of-sample testing shows they improve execution quality.

What we tried that doesn't work

We believe in showing failures alongside successes.

• Sell-the-top / fade spikes: Negative EV. Information-driven price moves don't mean-revert.
• Compression sniping: Negative on historical data, adverse selection live.
• Martingale / double-down on dips: Catastrophic — doubles position into adverse moves.
• Early NBA taker configs: Earlier versions underperformed until pacing/context features and stricter entry filters were added. Old model versions are not treated as current proof.
• Spread/Total mean-reversion exits: Spreads reprice permanently from score changes. Fix: hold to settlement instead.
• ESPN live WP as model input: Tested and rejected — creates circular dependency with market. ESPN pre-game WP works as a fixed prior.
• Conformal prediction intervals: Tested and rejected — added complexity without improving trading value.