Polymarket Ladder Esports Kills Trader

Trades monotonicity violations in esports Total Kills O/U market ladders on Polymarket. Each esports match spawns multiple kill total thresholds that must be...

Security Scan

The skill's code, instructions, and declared requirements are coherent with a Polymarket ladder-arbitrage trader and request only the single Simmer API key needed to place trades.

Purpose & Capability

Name/description (trade Polymarket kill-ladder violations) match the included code and metadata: the skill parses market questions, builds ladders, detects monotonicity violations, sizes orders, and submits trades via a SimmerClient. Required credential (SIMMER_API_KEY) and the declared pip dependency (simmer-sdk) are appropriate for a trading integration.

Instruction Scope

SKILL.md and trader.py limit behavior to market discovery, ladder construction, risk gating, and trade execution. The README emphasizes paper-by-default and requires an explicit --live flag for real USDC trades. I see no instructions to read unrelated system files, harvest other environment variables, or exfiltrate data to unknown endpoints.

Install Mechanism

No install script is present, but clawhub.json declares a pip dependency on 'simmer-sdk'. Installing a third-party Python package is a normal requirement for this functionality; pip packages carry moderate supply-chain risk compared to no-install skills. There are no downloads from untrusted URLs or archive extraction steps in the package metadata.

Credentials

Only a single high-value credential (SIMMER_API_KEY) is required, which is appropriate for a trading client. Tunables are exposed via environment variables but are non-secret and documented. The code reads only the declared SIMMER_* env vars and no unrelated secrets or config paths.

Persistence & Privilege

The skill is not always-enabled (always: false), autostart is false, and automaton runs default to paper mode. It does not request elevated system privileges or modify other skills' configuration. Autonomous invocation is allowed by platform default but presents no unusual additional privilege here.

Assessment

This skill appears coherent for its stated purpose, but take these precautions before enabling live trading: 1) Keep using paper mode to validate behavior and tune thresholds; 2) Verify the simmer-sdk package source (PyPI name and maintainer) before installing; 3) Treat SIMMER_API_KEY as high-value — store it in a secrets manager and rotate it if exposed; 4) Only run with --live after you've tested for false positives/negatives and reviewed logs; 5) Limit SIMMER API credentials to least privilege (trade scopes only) if the platform supports it.

Ladder — Esports Kills Curve Trader

This is a template. The default signal is kill total O/U ladder violation detection for esports markets -- remix it with additional titles, map types, or live match data feeds. The skill handles all the plumbing (market discovery, ladder construction, trade execution, safeguards). Your agent provides the alpha.

Strategy Overview

Esports matches (Dota 2, CS2, Overwatch) list multiple kill total thresholds for the same game:

• "Team A vs Team B Total Kills O/U 47.5" = 58%
• "Team A vs Team B Total Kills O/U 48.5" = 55%
• "Team A vs Team B Total Kills O/U 49.5" = 52%
• "Team A vs Team B Total Kills O/U 50.5" = 54% <-- VIOLATION

These form a kill total ladder -- higher thresholds must always have lower OVER probability. When they don't, it is structural arbitrage.

The Edge: Kill Ladder Monotonicity Arbitrage

The probability of going OVER a lower kill threshold must always be greater than or equal to going OVER a higher threshold:

P(O/U 47.5 OVER) >= P(O/U 48.5 OVER) >= P(O/U 49.5 OVER) >= P(O/U 50.5 OVER)

If a higher threshold is priced above a lower one, the ladder is broken -- pure structural arbitrage.

Why Esports Kill Markets Are Especially Mispriced

1. Narrow thresholds -- Dota 2 kill ladders use 1.0 increments (47.5, 48.5, 49.5) creating many adjacent rungs where violations can form
2. Retail trades in silos -- most users view each O/U line independently, not the full ladder
3. No market maker enforcement -- unlike sportsbooks, Polymarket has no central entity maintaining ladder consistency
4. Game-specific volatility -- Dota 2 average kills per game is ~55 with high variance, meaning the ladder spans a wide range of plausible outcomes
5. Multi-game series -- BO3 matches create separate ladders for Game 1, Game 2, Game 3, tripling the surface area for violations

Signal Logic

1. Discover esports kills O/U markets via keyword search + bulk market fetch fallback
2. Parse each question: extract match_key, game_number, kill_threshold
3. Filter non-esports markets with regex (remove crypto, politics, weather noise)
4. Group into ladders by (match_key, game_number)
5. For each ladder with 2+ points:
   • Sort by kill threshold ascending
   • Check monotonicity: P(O/U X OVER) must decrease as X increases
6. Rank violations by magnitude
7. Trade only violations that pass threshold gates (YES_THRESHOLD / NO_THRESHOLD)
8. Size by conviction (violation magnitude), not flat amount

Remix Signal Ideas

• Liquipedia API: Pull team Elo ratings and recent match history — teams with high-variance playstyles create wider kill distributions, meaning more O/U lines and more violation opportunities
• Dotabuff / OpenDota API: Fetch hero pick/ban data and lane matchups for the specific game — aggressive drafts with teamfight heroes predict higher kill totals, letting you weight which side of the monotonicity break to trade
• Live match data (PandaScore API): Stream real-time kill counts during a BO3 — if Game 1 had 60 kills, Game 2 O/U lines should shift up; if the market hasn't adjusted, that's an additional timing edge
• Map/patch meta analysis: After major game patches, kill totals shift systematically — track patch notes and historical post-patch kill averages to predict which direction the curve should move

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	0.03	Min ladder violation magnitude to trigger a trade

Edge Thesis

Traditional sportsbooks have professional line-setters who enforce consistency across O/U lines for the same game. Polymarket has no such mechanism -- each kill total O/U market is priced by its own order book with its own liquidity pool. Esports kill markets are particularly vulnerable because:

• Kill thresholds are tightly spaced (1.0 increments), creating many adjacent rungs
• Multi-game series (BO3/BO5) multiply the number of ladders per match
• Esports-specific liquidity is thinner than mainstream sports, widening violations
• New thresholds added mid-match don't inherit pricing from neighbors
• Large directional flow on one threshold doesn't propagate to adjacent ones

This skill treats the kill total O/U ladder as a monotonic probability curve and trades the repair.

Dependency

simmer-sdk by Simmer Markets (SpartanLabsXyz)

• PyPI: https://pypi.org/project/simmer-sdk/
• GitHub: https://github.com/SpartanLabsXyz/simmer-sdk