Polymarket Ladder Nhl Hockey Trader
v0.0.3Trades monotonicity violations in NHL hockey O/U market ladders and spread-vs-total consistency on Polymarket. Each game spawns multiple O/U lines that must...
Ladder — NHL Hockey Curve Trader
This is a template. The default signal detects monotonicity violations in NHL hockey O/U ladders and spread-vs-total inconsistencies -- remix it with additional leagues, period-based markets, or live odds feeds. The skill handles all the plumbing (market discovery, ladder grouping, monotonicity checks, spread consistency, trade execution, safeguards). Your agent provides the alpha.
Strategy Overview
Polymarket lists multiple O/U lines for each NHL game at different totals:
- Wild vs Bruins: O/U 4.5 = 62%, O/U 5.5 = 45%, O/U 7.5 = 12%
- Maple Leafs vs Blues: O/U 4.5 = 65%, O/U 5.5 = 50%, O/U 6.5 = 30%, O/U 7.5 = 15%
- Spread markets: "Spread: Wild (-1.5)" = 42%
These O/U markets form a ladder that must be monotonically decreasing:
P(O/U 4.5 OVER) >= P(O/U 5.5 OVER) >= P(O/U 6.5 OVER) >= P(O/U 7.5 OVER)
Additionally, spread markets constrain the O/U distribution -- a large spread (one team heavily favored) implies a scoring profile that must be consistent with the total.
This skill reconstructs each game's full ladder and trades where it is mathematically broken.
The Edge: NHL Ladder Arbitrage
Violation Type 1: O/U Monotonicity
Within the same game, the probability of going OVER must decrease as the line increases. If a higher line is priced above a lower line, the curve is broken -- pure structural arbitrage:
If P(O/U 5.5 OVER) > P(O/U 4.5 OVER):
Buy YES on O/U 4.5 (underpriced)
Buy NO on O/U 5.5 (overpriced)
This also checks non-adjacent pairs (e.g., O/U 7.5 vs O/U 4.5) for larger violations that span multiple rungs of the ladder.
Violation Type 2: Spread-vs-Total Consistency
A large spread implies one team is heavily favored, which shifts the expected scoring distribution:
- If spread >= 2.5 goals, the game is expected to be high-scoring, so P(O/U 5.5 OVER) should be at least 40%
- If spread >= 1.5 goals, P(O/U 4.5 OVER) should be at least 50%
When the O/U curve is inconsistent with the spread, the total is likely mispriced.
Why This Works
- Retail trades in silos -- most users bet on individual O/U lines without cross-referencing the full ladder
- No market maker enforcement -- unlike sportsbooks, there is no central entity maintaining consistency across O/U lines for the same game
- Mathematical, not opinion -- monotonicity violations are provable inconsistencies in the implied probability distribution
- NHL-specific density -- NHL games generate multiple O/U lines (4.5, 5.5, 6.5, 7.5) plus spread markets, creating a rich surface for inconsistencies
- Injury and lineup news -- goalie changes and late scratches move individual lines without propagating to the full ladder
Signal Logic
- Discover all NHL-related markets via keyword search (team names, "NHL", "hockey", "O/U")
- Fallback to
get_markets(limit=200)if keyword search yields few results - Parse each question: extract (game_key, ou_line) for O/U markets; (game_key, spread_team, spread_value) for spread markets
- Filter with NHL team regex, exclude non-hockey matches
- Group O/U markets into ladders by game_key
- Match spread markets to game ladders by team name overlap
- For each game ladder with 2+ O/U markets:
- Check monotonicity: P(OVER) must decrease as line increases
- Check non-adjacent pairs for larger violations
- For games with spread + O/U markets:
- Check spread-vs-total consistency
- Rank violations by magnitude
- Trade only violations that also pass threshold gates (
YES_THRESHOLD/NO_THRESHOLD) - Size by conviction, not flat amount
Remix Signal Ideas
- NHL API live stats: Use the NHL API (api-web.nhle.com) for live game stats such as shot attempts, Corsi, and Fenwick ratings to predict whether a game is trending over or under before the O/U lines adjust
- Expected goals (xG) model: Incorporate MoneyPuck.com expected goals data to build an independent estimate of game totals and compare against Polymarket's implied distribution
- Historical O/U line movement: Track historical over/under line movement data from sportsbook APIs to detect when Polymarket's O/U ladder lags behind sharp market moves
- Weather and arena conditions: Factor in outdoor game conditions (Winter Classic, Stadium Series), altitude (Denver), and ice quality to adjust scoring expectations -- outdoor games and altitude historically produce different scoring profiles
- Back-to-back scheduling fatigue: Use the NHL schedule to identify teams playing on back-to-back nights, which correlates with weaker goaltending and higher totals -- trade the O/U ladder when fatigue is not yet priced in
Safety & Execution Mode
The skill defaults to paper trading (venue="sim"). Real trades only with --live flag.
| Scenario | Mode | Financial risk |
|---|---|---|
python trader.py | Paper (sim) | None |
| Cron / automaton | Paper (sim) | None |
python trader.py --live | Live (polymarket) | Real USDC |
autostart: false and cron: null mean nothing runs automatically until configured in Simmer UI.
Required Credentials
| Variable | Required | Notes |
|---|---|---|
SIMMER_API_KEY | Yes | Trading authority. Treat as a high-value credential. |
Tunables (Risk Parameters)
All declared as tunables in clawhub.json and adjustable from the Simmer UI.
| Variable | Default | Purpose |
|---|---|---|
SIMMER_MAX_POSITION | 40 | Max USDC per trade at full conviction |
SIMMER_MIN_TRADE | 5 | Floor for any trade |
SIMMER_MIN_VOLUME | 5000 | Min market volume filter (USD) |
SIMMER_MAX_SPREAD | 0.08 | Max bid-ask spread |
SIMMER_MIN_DAYS | 0 | Min days until resolution (0 = allow same-day) |
SIMMER_MAX_POSITIONS | 8 | Max concurrent open positions |
SIMMER_YES_THRESHOLD | 0.38 | Buy YES only if market probability <= this |
SIMMER_NO_THRESHOLD | 0.62 | Sell NO only if market probability >= this |
SIMMER_MIN_VIOLATION | 0.05 | Min ladder violation magnitude to trigger a trade |
Edge Thesis
Traditional sportsbooks have professional line-setters who enforce consistency across all O/U lines for the same game. Polymarket has no such mechanism -- each O/U market (4.5, 5.5, 6.5, 7.5) is priced by its own order book with its own liquidity pool. This creates systematic micro-inconsistencies in the implied scoring distribution, especially when:
- Goalie changes or injury news moves one O/U line but not others
- New O/U lines are added at different totals without re-pricing existing ones
- Large directional flow on one line does not propagate to adjacent lines
- Spread markets move on team news but the O/U ladder does not adjust
This skill treats the full O/U ladder as a consistency curve and trades the repair.
Dependency
simmer-sdk by Simmer Markets (SpartanLabsXyz)