This is a template.
The default signal trades cascade lag from nearly-resolved trigger events to downstream target markets -- remix it with real-time news feeds, order-flow data, or cross-venue latency analysis.
The skill handles all the plumbing (market discovery, trade execution, safeguards). Your agent provides the alpha.
策略概述
当 Polymarket 上的重大事件解决时，重新定价不会立即发生。一阶效应（直接相关的市场）在几分钟内重新定价。但二阶和三阶效应——通过经济或地缘政治传播渠道连接的市场——需要数小时才能完全调整。
级联示例：伊朗停火协议解决 YES →
- 一阶效应（即时）： 地缘政治市场重新定价
- 二阶效应（30分钟-2小时）： 石油下跌，BTC 上涨
- 三阶效应（2小时-8小时）： 股票/衰退市场上涨
本技能扫描所有市场，识别即将解决的事件（p > 90% 或 p < 10%，或在 24 小时内解决），映射预期的级联链，并交易尚未重新定价的下游目标。
边缘论点
预测市场参与者是专家。交易伊朗停火市场的人与交易 BTC 或石油市场的人不是同一个人。跨资产重新定价需要信息在孤立的参与者池之间传播。这创造了结构性滞后，在三阶效应和 off-hours 期间最大，当时活跃的参与者较少。
边缘不在于预测触发事件。边缘在于比跨资产信息流更快。
信号逻辑
1. 触发器检测
扫描所有市场并识别即将解决的事件：
- 概率 > 90% 或 < 10%（市场已基本决定）
- 24 小时内解决且概率 > 50% 或 < 50%
2. 事件分类
将每个触发器分类到类别：地缘政治、加密货币、大宗商品、健康、天气、股票。对于地缘政治，进一步分类为升级或缓和。
3. 级联链映射
预定义的级联链将触发器信号映射到预期的目标变动：
| 触发器 | 二阶效应 | 三阶效应 |
|---|
| 地缘政治升级 YES | 石油 UP，加密货币 DOWN | 股票 DOWN |
| 地缘政治缓和 YES | 石油 DOWN，加密货币 UP | 股票 UP |
| 加密货币里程碑 UP | 股票 UP |
| 极端天气 YES | 大宗商品 UP |
| 健康爆发 YES | 股票 DOWN，大宗商品 UP |
4. 滞后检测
对于每个级联目标，测量它在预期方向上移动了多少。
一个处于 35% 而级联表示应该 UP 的目标的滞后为 15%（从 50% 中点）。
仅在滞后超过 CASCADE_LAG 阈值时交易目标。
5. 信念式 sizing
所有交易使用标准信念公式：
- YES：
信念 = (YES_THRESHOLD - p) / YES_THRESHOLD，由级联滞后增强
- NO：
信念 = (p - NO_THRESHOLD) / (1 - NO_THRESHOLD)，由级联滞后增强
- 规模：
max(MIN_TRADE, 信念 * MAX_POSITION)
混音创意
- 添加实时新闻 API 以在交易级联前确认触发器事件
- 按触发器市场成交量加权级联强度（更大的触发器 = 更强的级联）
- 添加时间衰减：级联信号随着触发器检测后的小时数增加而减弱
- 跨场所套利：检查 Kalshi/Metaculus 以确认级联滞后
- 添加订单簿深度分析以估计剩余的重新定价量
安全与执行模式
该技能默认使用模拟交易（venue="sim"）。仅在使用 --live 标志时进行真实交易。
| 场景 | 模式 | 财务风险 |
|---|
python trader.py | 模拟（sim） | 无 |
| Cron / automaton | 模拟（sim） | 无 |
python trader.py --live | 真实（polymarket） | 真实 USDC |
autostart: false 和
cron: null 意味着在 Simmer UI 中配置之前不会自动运行任何内容。
所需凭证
| 变量 | 必需 | 备注 |
|---|
SIMMER_API_KEY | 是 | 交易授权。视为高价值凭证。 |
可调参数（风险参数）
所有在 clawhub.json 中声明为 tunables，可从 Simmer UI 调整。
| 变量 | 默认值 | 用途 |
|---|
SIMMER_MAX_POSITION | 40 | 满信念时每笔交易的最大 USDC |
SIMMER_MIN_TRADE | 5 | 任何交易的最低金额 |
SIMMER_MIN_VOLUME | 10000 | 最小市场成交量过滤器（USD） |
SIMMER_MAX_SPREAD | 0.07 | 最大买卖价差 |
SIMMER_MIN_DAYS | 3 | 距离解决的最小天数 |
SIMMER_MAX_POSITIONS | 6 | 最大并发未平仓头寸 |
SIMMER_YES_THRESHOLD | 0.38 | 仅当市场概率 <= 此值时买入 YES |
SIMMER_NO_THRESHOLD | 0.62 | 仅当市场概率 >= 此值时卖出 NO |
SIMMER_CASCADE_LAG | 0.08 | 交易级联目标前的最小滞后（从 50% 中点） |
依赖项
simmer-sdk by Simmer Markets (SpartanLabsXyz)
- PyPI: https://pypi.org/project/simmer-sdk/
- GitHub: https://github.com/SpartanLabsXyz/simmer-sdk
This is a template.
The default signal trades cascade lag from nearly-resolved trigger events to downstream target markets -- remix it with real-time news feeds, order-flow data, or cross-venue latency analysis.
The skill handles all the plumbing (market discovery, trade execution, safeguards). Your agent provides the alpha.
Strategy Overview
When a major event resolves on Polymarket, the repricing does not happen all at once. First-order effects (the directly related markets) reprice within minutes. But second and third-order effects -- the markets connected through economic or geopolitical transmission channels -- take hours to fully adjust.
Example cascade: Iran ceasefire resolves YES ->
- 1st order (immediate): Geopolitics markets reprice
- 2nd order (30min-2h): Oil drops, BTC rises
- 3rd order (2h-8h): Equity/recession markets rise
This skill scans all markets, identifies triggers that are nearly resolved (p > 90% or p < 10%, or resolving within 24h), maps the expected cascade chain, and trades the downstream targets that have NOT yet repriced.
Edge Thesis
Prediction market participants are specialists. The person trading Iran ceasefire markets is not the same person trading BTC or oil markets. Cross-asset repricing requires information to travel between siloed participant pools. This creates a structural lag that is largest for 3rd-order effects and during off-hours when fewer participants are active.
The edge is not in predicting the trigger event. The edge is in being faster than cross-asset information flow.
Signal Logic
1. Trigger Detection
Scan all markets and identify nearly-resolved events:
- Probability > 90% or < 10% (market has effectively decided)
- Resolution within 24 hours and probability > 50% or < 50%
2. Event Classification
Classify each trigger into a category: geopolitics, crypto, commodity, health, weather, equity. For geopolitics, further classify as escalation or de-escalation.
3. Cascade Chain Mapping
Predefined cascade chains map trigger signals to expected target movements:
| Trigger | 2nd Order | 3rd Order |
|---|
| Geo escalation YES | Oil UP, Crypto DOWN | Equity DOWN |
| Geo de-escalation YES | Oil DOWN, Crypto UP | Equity UP |
| Crypto milestone UP | Equity UP |
| Weather extreme YES | Commodity UP |
| Health outbreak YES | Equity DOWN, Commodity UP |
4. Lag Detection
For each cascade target, measure how much it has NOT moved in the expected direction. A target sitting at 35% when the cascade says it should go UP has a lag of 15% (from the 50% midpoint). Only trade targets where lag exceeds CASCADE_LAG threshold.
5. Conviction-Based Sizing
All trades use the standard conviction formula:
- YES:
conviction = (YES_THRESHOLD - p) / YES_THRESHOLD, boosted by cascade lag
- NO:
conviction = (p - NO_THRESHOLD) / (1 - NO_THRESHOLD), boosted by cascade lag
- Size:
max(MIN_TRADE, conviction * MAX_POSITION)
Remix Ideas
- Add real-time news API to confirm trigger events before trading cascade
- Weight cascade strength by trigger market volume (bigger trigger = stronger cascade)
- Add time-decay: cascade signal weakens as hours pass since trigger detection
- Cross-venue arbitrage: check Kalshi/Metaculus for cascade lag confirmation
- Add order-book depth analysis to estimate how much repricing is left
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 | 10000 | Min market volume filter (USD) |
SIMMER_MAX_SPREAD | 0.07 | Max bid-ask spread |
SIMMER_MIN_DAYS | 3 | Min days until resolution |
SIMMER_MAX_POSITIONS | 6 | Max concurrent open positions |
SIMMER_YES_THRESHOLD | 0.38 | Buy YES only if market probability <= this value |
SIMMER_NO_THRESHOLD | 0.62 | Sell NO only if market probability >= this value |
SIMMER_CASCADE_LAG | 0.08 | Min lag (from 50% midpoint) before trading a cascade target |
Dependency
simmer-sdk by Simmer Markets (SpartanLabsXyz)
- PyPI: https://pypi.org/project/simmer-sdk/
- GitHub: https://github.com/SpartanLabsXyz/simmer-sdk
by