House model: NHL totals
Paper bot · Retired
Bets only when the model and market meet its published selection policy.
Historical record unavailable
Live record lost in the 24 Oct 2026 server move.
Simulated paper trading only — orders are not sent to Polymarket.
The tested policy is off. The picks are still published.
The backtest found no edge worth taking, so this bot places no bets. The model still publishes and grades its picks.
Places no bets: the backtest found no edge worth taking. Picks are still published and graded.
Test ROI −11.7% across 123 test bets. The 95% interval is −28.8% to +5.9%. That range includes losing money.
All-edge baseline ROI −15.2%.
Backtest prices: Polymarket price at kickoff − 3 h, 2025-26 (+0.5c slippage, taker fee rate 0.05). A backtest uses historical prices; the paper record above uses live quotes.
The model publishes a chance for every game and is graded on every game, whether or not this policy places a bet. A published pick is a prediction; a paper bet is an execution. This policy places no paper bets.
Every published pick and its grade →Show policy methodologyHide policy methodology
Tested on Polymarket's own NHL prices from 2025-26, buying 3 hours before the start and paying the fee. The old rule (bet every over/under where the model beats the price by 4c) returned -19% on the first 60% of the season (82 bets) and -15% on the last 40% (71 bets). The best filter on the first 60% (26 bets, +10%) returned -8% on the rest (17 bets). No filter held up out of sample, and the market priced these games at least as well as the model, so the bot places no bets. Picks are still published and graded for the record.
- Periods
- train base model: seasons before 2025-26, no prices · validate selection tuned on 2025-10-08 to 2026-01-28 (first 60% of 2025-26; no separate validation season exists) · test 2026-01-29 to 2026-06-14
- Prices
- Polymarket price at kickoff − 3 h, 2025-26 (+0.5c slippage, taker fee rate 0.05)
- Test bets
- 123 · hit 44.7%
- Test ROI
- −11.7% (95% CI −28.8% to +5.9%)
- Every ≥4¢ edge
- −15.2%
- Version
- 2026-10-02
In plain words
TL;DR
Off: the goals model publishes picks on NHL over/unders, but the bot places no bets because the backtest found no edge after fees.
How it bets
The model estimates how many goals both teams score together and the chance the total goes over each line. Polymarket lists several lines per game; the bot only trades the main one (priced closest to 50¢), once per game; then see "Why 4¢" below.
Why this bot is off
- Before each game the model publishes its chance for every side, next to the market's price.
- In the backtest, betting these picks lost money once the bookmaker's or Polymarket's cut is paid.
- So the bot sits out. Not betting is a result too: the market priced these games better than the model.
Exact rules
Off: the backtest found no edge worth betting, so this bot places no bets. The model still publishes a pick for every game so its record stays public. Tested on Polymarket's own NHL prices from 2025-26, buying 3 hours before the start and paying the fee. The old rule (bet every over/under where the model beats the price by 4c) returned -19% on the first 60% of the season (82 bets) and -15% on the last 40% (71 bets). The best filter on the first 60% (26 bets, +10%) returned -8% on the rest (17 bets). No filter held up out of sample, and the market priced these games at least as well as the model, so the bot places no bets. Picks are still published and graded for the record.
Bets
No published bets yet: this record has no settled, closed or open bets.