← All posts

Backtesting Articles

14 posts tagged “backtesting”.

polymarket kalshi market-microstructure

Where to Get Historical Polymarket & Kalshi Order-Book Data (2026): An Honest Comparison

Vendor-neutral 2026 buyer's guide to historical Polymarket and Kalshi order-book data: capture it yourself or buy — what each vendor has, what it costs.
prediction-markets calibration polymarket

Are Prediction Markets Actually Calibrated? We Measured It — Then Checked Our Own Bot

When a prediction market prices a team at 87%, do those teams really win ~87% of the time? Using live-captured, outcome-labeled MLB books from Polymarket...
polymarket kalshi market-microstructure

How to Get Historical Polymarket Order Book Data

Polymarket's API gives you trades but not historical depth, and the book can't be rebuilt after the fact. Your real options for order-book history.
polymarket backtesting historical-data

Backtesting Polymarket Strategies: Tools, Datasets, and the Depth Data Problem

Compare Polymarket price history, recorded book quotes and L2 depth for backtests. Check clocks, costs and capture gaps before choosing a dataset.
polymarket backtesting strategy

Polymarket Paper Trading: How to Test Strategies Without Risking Money in 2026

Polymarket has no built-in paper trading mode. Here are the real options for testing strategies without risk — Manifold Markets, historical price...
nhl backtesting calibration

We Backtested Our NHL Model on the 2025 Stanley Cup Playoffs. It Hit 60.5% Across 86 Games.

Our NHL model replayed the 2025 Stanley Cup playoffs on pre-playoff data: 52 of 86 games correct (60.5%), with the Cup Final correctly read as a coin flip.
mlb backtesting calibration

We Backtested Our MLB Model on the 2025 Postseason. It Hit 59.6% Across 47 Games.

Our MLB model replayed the entire 2025 postseason on pre-playoff data: 28 of 47 correct (59.6%). Full breakdown, including every missed ALCS pick.
nfl backtesting calibration

We Backtested Our NFL Model on the 2026 Playoffs. It Called Super Bowl LX Correctly.

Our NFL model replayed the 2025-26 playoffs on pre-playoff data only: 9 of 13 correct (69.2%), including Seattle over New England in Super Bowl LX.
nba backtesting calibration

We Backtested Our NBA Model on the 2025 Playoffs. It Called OKC Correctly All Season.

After the 2025 NBA Finals wrapped with OKC beating Indiana in seven games, we ran our NBA win-probability model on all 84 postseason games using only...
ncaaf calibration machine-learning

Why Our First CFB Model Lost Money (And the One Feature That Fixed It)

Our first college football model backtested at -2.1c/trade over 728 simulated positions. The missing feature, the diagnosis, and the corrected numbers.
ncaamb results calibration

Our NCAAMB Model's 2025-26 Season Report: 5,345 Games Backtested, 68.2% Accuracy, ECE 4.39%

Our college basketball model backtested on all 5,345 non-tournament games of 2025-26: 68.2% accuracy, Brier 0.2064, ECE 4.39%, full reliability table.
ncaamb backtesting calibration

We Backtested Our Model on the 2026 March Madness Bracket. It Hit 71.6%.

Our NCAAMB model replayed all 67 games of March Madness 2026 on pre-tournament data: 48 correct (71.6%), 3-for-3 on the Final Four. Every miss listed.
trading market-microstructure backtesting

99% of Your Backtested Edge Doesn't Exist: Execution Quality on Prediction Markets

We backtested 237 trade signals with and without execution constraints. 99% of theoretical profit vanished. Here's what actually kills your trades — and...
backtesting python strategy

How to Backtest a Sports Betting Strategy in Python (Without Fooling Yourself)

A practical guide to backtesting sports betting and prediction market strategies in Python. Covers the common pitfalls — survivorship bias, look-ahead...