Polymarket Bundle Crypto Fade Trader

Fades strong directional crypto moves on Polymarket 5-minute interval markets. After 3+ consecutive high-conviction same-direction intervals (>58% probability), the next interval tends to mean-revert -- a well-documented microstructure effect in crypto. Targets BTC, ETH, and SOL Up or Down bundles with conviction-based position sizing scaled by streak strength.

Bundle -- Crypto Momentum Fade Trader

This is a template. The default signal detects strong directional streaks in crypto 5-minute interval markets and fades them using conviction-based sizing. The skill handles all the plumbing (interval parsing, streak detection, trade execution, safeguards). Your agent provides the alpha.

Strategy Overview

Polymarket lists 5-minute interval markets for BTC, ETH, and SOL: "Will BTC be Up or Down in the 14:00-14:05 ET interval?" These resolve to YES (up) or NO (down) based on the actual price movement. When a coin moves sharply in one direction over 15-30 minutes -- 3 to 6 consecutive strong same-direction intervals at >58% probability -- the NEXT interval tends to mean-revert. This is the "momentum fade," a well-documented microstructure effect in crypto markets where short-term directional momentum exhausts itself and reverses.

Edge

Unlike a generic streak trader that counts any streak of >50% intervals, this skill specifically targets strong directional moves where each interval in the streak exceeds the FADE_THRESHOLD (default 58%). This filters out noise and focuses on genuine momentum exhaustion events. The fade is structurally sound because:

1. Microstructure mean-reversion -- short-horizon crypto returns exhibit negative autocorrelation after extended directional moves, documented across BTC, ETH, and SOL on 5-minute timeframes
2. Retail momentum chasing -- Polymarket participants observe a streak and price the next interval as a continuation, creating a systematic overpricing of momentum
3. Arbitrage-free pricing constraint -- the 5-minute return distribution has bounded variance; after 3+ consecutive strong-direction intervals, the conditional probability of continuation drops below the unconditional probability
4. Bundle structure -- each interval resolves independently, so a streak of 3 strong-up intervals does not mechanically cause the 4th to also be up

Signal Logic

1. Discover crypto interval markets via keyword search (Bitcoin Up or Down, BTC Up or Down, Ethereum Up or Down, Solana Up or Down) with a get_markets(limit=200) fallback
2. Parse each question to extract (coin, date, start_time, end_time) using regex
3. Group intervals by (coin, date) and sort by time
4. Scan each group for streaks of FADE_LENGTH+ consecutive intervals where ALL have probability >= FADE_THRESHOLD (strong-up) or <= 1 - FADE_THRESHOLD (strong-down)
5. After detecting a strong move, check the NEXT interval:
   • Strong-up streak -> next interval should mean-revert down; if still priced >= NO_THRESHOLD, sell NO
   • Strong-down streak -> next interval should mean-revert up; if still priced <= YES_THRESHOLD, buy YES
6. Size by conviction (distance from threshold), not flat amount

Remix Signal Ideas

• Binance/Coinbase websocket -- feed real-time order flow data to detect momentum exhaustion before the 5-minute interval resolves; trade the interval market while the fade is still mispriced
• Funding rate signal -- when perpetual funding rates are extremely positive/negative, the fade has higher expected value because leveraged positions are being unwound
• Volume profile -- weight the streak detection by actual trading volume in each interval; high-volume directional moves exhaust faster than low-volume drifts
• Cross-coin correlation -- if BTC has a strong-up streak and ETH/SOL have not yet followed, the fade on BTC is weaker (genuine broad move) vs. BTC-only streaks (more likely noise)
• Volatility regime -- adjust FADE_THRESHOLD dynamically based on realised volatility; in high-vol regimes, require stronger streaks before fading

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 (min USDC regardless of conviction)
SIMMER_MIN_VOLUME	3000	Min market volume filter (USD)
SIMMER_MAX_SPREAD	0.10	Max bid-ask spread
SIMMER_MIN_DAYS	1	Min days until resolution
SIMMER_MAX_POSITIONS	10	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_FADE_THRESHOLD	0.58	Min probability for a "strong" interval in streak detection
SIMMER_FADE_LENGTH	3	Min consecutive strong intervals to trigger a fade

Edge Thesis

Crypto 5-minute interval markets on Polymarket exhibit momentum chasing by retail participants. When a coin moves sharply in one direction -- 3+ consecutive intervals where each is priced at >58% in the same direction -- the pricing of the next interval systematically overestimates continuation probability. This is a direct consequence of short-horizon mean-reversion in crypto microstructure: after extended directional moves, the conditional probability of further continuation drops below the market-implied probability. The fade exploits this gap with conviction-based sizing that scales with the distance from the trading threshold.

Dependency

simmer-sdk by Simmer Markets (SpartanLabsXyz)

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