shield_lock
金大哥
← 返回 Skill 列表
extension
分类: 其它需要 API Key

Polymarket Ladder F1 Championship Trader

在 Polymarket 上交易 F1 冠军市场的分布总额违规。车手获胜概率形成必须总和约 100% 的分布 —...

person作者: diagnostikonhubclawhub
open_in_new仓库download下载资源包

Ladder -- F1 Championship Distribution Trader

This is a template. The default signal exploits distribution-sum violations in F1 championship winner markets. No external API required beyond simmer-sdk. The skill discovers F1 driver championship markets, groups them by championship, checks whether the implied probability distribution sums to ~100%, and trades the correction when it drifts.

Strategy Overview

In a winner-takes-all market like "Who will be the 2026 F1 Drivers' Champion?", the individual driver probabilities must sum to approximately 100%. On Polymarket, each driver is listed as a separate binary market. Because each market has its own bid-ask spread, retail flow, and narrative momentum, the implied probabilities frequently drift apart:

  • Sum > 105%: The field is collectively overpriced. Each driver's YES price is inflated relative to the fair share of 100%. We sell the most overpriced drivers (buy NO on the highest-probability ones).
  • Sum < 95%: The field is collectively underpriced. Each driver's YES price is too low. We buy the most underpriced drivers (buy YES on the lowest-probability ones).

This is a structural arbitrage on probability distributions, not a directional bet on any single driver winning or losing.

The Core Insight: Why Distribution Violations Exist

Three mechanisms create and sustain distribution-sum violations:

1. Independent bid-ask spreads Each driver market has its own spread. When you sum 20 drivers' mid prices, the individual spreads compound. A 2-3 cent spread on each of 20 markets can push the total sum 10-15% above or below 100%.

2. Narrative-driven retail flow When news breaks (e.g., Verstappen dominates a qualifying session), retail piles into that driver's YES market, pushing the entire field's sum above 100%. The other drivers' prices don't drop fast enough to compensate because there's no automated arbitrage mechanism across separate binary markets.

3. Stale pricing on low-liquidity drivers The top 3-5 drivers have active order books. The remaining 15+ drivers may have stale prices that don't adjust when the favorites move, creating persistent sum violations.

Signal Logic

Step 1 -- Market Discovery

Keyword search across F1-specific terms plus a get_markets(limit=200) fallback. All candidates are filtered by a regex that matches F1/motorsport market questions.

Step 2 -- Championship Grouping

Markets are parsed to extract (championship_name, driver_name) tuples using regex patterns that match:

  • "Will Verstappen be the 2026 F1 Drivers' Champion?"
  • "Will Gasly win the 2026 Formula 1 championship?"

Markets are grouped by championship name.

Step 3 -- Distribution Sum Check

For each championship group, sum all driver probabilities. If |sum - 1.0| < MIN_VIOLATION (default 5%), the distribution is within tolerance and we skip the group.

Step 4 -- Signal Direction

  • Overpriced field (sum > 105%): Sort drivers by probability descending. Buy NO on drivers above NO_THRESHOLD (most overpriced first).
  • Underpriced field (sum < 95%): Sort drivers by probability ascending. Buy YES on drivers below YES_THRESHOLD (most underpriced first).

Step 5 -- Conviction-Based Sizing

For NO (overpriced field):

conviction = (p - NO_THRESHOLD) / (1 - NO_THRESHOLD)
violation_boost = min(1.5, 1.0 + violation_magnitude)
size = max(MIN_TRADE, conviction * violation_boost * MAX_POSITION)

For YES (underpriced field):

conviction = (YES_THRESHOLD - p) / YES_THRESHOLD
violation_boost = min(1.5, 1.0 + violation_magnitude)
size = max(MIN_TRADE, conviction * violation_boost * MAX_POSITION)

The violation boost increases position size when the distribution-sum violation is larger (more edge).

How Sizing Works

With defaults (YES_THRESHOLD=38%, NO_THRESHOLD=62%, MIN_TRADE=$5, MAX_POSITION=$40, violation=10%):

Overpriced field (buy NO on high-p drivers):

| Driver p | Conviction | Violation boost | Size | |---|---|---|---| | 62% (at threshold) | 0% | 1.10x | $5 (floor) | | 70% | 21% | 1.10x | $9 | | 80% | 47% | 1.10x | $21 | | 90% | 74% | 1.10x | $33 |

Underpriced field (buy YES on low-p drivers):

| Driver p | Conviction | Violation boost | Size | |---|---|---|---| | 38% (at threshold) | 0% | 1.10x | $5 (floor) | | 30% | 21% | 1.10x | $9 | | 20% | 47% | 1.10x | $21 | | 10% | 74% | 1.10x | $33 |

Keywords Monitored

F1, Formula 1, champion, Drivers Champion, Grand Prix,
Verstappen, Hamilton, Norris, Leclerc, Piastri,
Russell, Gasly, Albon, Bottas

Remix Signal Ideas

  • Cross-championship arbitrage: Compare distribution-sum violations across multiple championships (F1, MotoGP, IndyCar) and prioritize the most violated distribution
  • Historical sum tracking: Track the distribution sum over time and trade mean-reversion when the sum spikes or dips
  • News-driven violation detection: Wire in an F1 news feed and trigger scans immediately after major race results or qualifying sessions, when retail flow is most likely to create violations
  • Team-pair arbitrage: Within a team, the two drivers' probabilities should roughly reflect their relative performance -- trade when team-pair ratios diverge from season data
  • Volume-weighted conviction: Weight conviction by each driver market's volume -- low-volume markets have staler prices and potentially more edge

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 -- nothing runs automatically until you configure it in Simmer UI.

Required Credentials

| Variable | Required | Notes | |---|---|---| | SIMMER_API_KEY | Yes | Trading authority. Treat as 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 (reached at 100% conviction) | | SIMMER_MIN_VOLUME | 5000 | Min market volume filter | | SIMMER_MAX_SPREAD | 0.06 | Max bid-ask spread (6%) | | SIMMER_MIN_DAYS | 7 | Min days until resolution | | SIMMER_MAX_POSITIONS | 8 | Max concurrent open positions | | SIMMER_YES_THRESHOLD | 0.38 | Buy YES when driver probability <= this in underpriced field | | SIMMER_NO_THRESHOLD | 0.62 | Buy NO when driver probability >= this in overpriced field | | SIMMER_MIN_TRADE | 5 | Floor for any trade (min USDC regardless of conviction) | | SIMMER_MIN_VIOLATION | 0.05 | Min distribution-sum violation to trigger trades (5%) |

Edge Thesis

F1 driver championship winner probabilities on Polymarket systematically deviate from a 100% total because of three structural mechanisms:

  1. Retail trades drivers individually -- users buy YES on their favorite driver without checking whether the field total already exceeds 100%. A wave of enthusiasm after a race result inflates the winner without deflating the rest.
  2. No cross-market consistency enforcement -- unlike a traditional bookmaker who adjusts the entire field when one runner's odds move, Polymarket has no market maker enforcing that the sum of all driver probabilities stays at 100%. Each driver is a separate order book with independent liquidity.
  3. Driver entry and exit creates transient imbalances -- when a new driver is added to the championship market (mid-season replacement, rookie promotion) or a driver is removed (retirement, disqualification), the existing prices do not instantly re-normalize. The sum drifts until enough traders notice and correct it.

These three forces guarantee recurring distribution-sum violations that revert as resolution approaches and the market must price a single winner at 100%.

Dependency

simmer-sdk by Simmer Markets (SpartanLabsXyz)

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