Polytutor

Round 5, 27/39: CLXM. Projects Binance velocity across an assumed Chainlink lag, compares it with the price-to-beat, one Kelly-sized buy in the last ~2s. Live-ops tell it's dead: high-confidence signals stop beating low-confidence ones. Research only. github.com/Polytutor-AI-Labs…

Round 5, 26/39: CLPTB. In the last seconds, buy the side of the price-to-beat Chainlink is on, allow one opposite re-entry if it crosses, hold to settlement. Live-ops tell it's dead: re-entries keep landing just before price whips back. Research only. github.com/Polytutor-AI-Labs…

Round 5, 25/39: CLCBM. Adjust Chainlink by Coinbase's last ~2s move, see which side of the price-to-beat the estimate lands on, buy it and hold to settlement. Live-ops tell it's dead: by the time the estimate flips, the ask already has. Research only. github.com/Polytutor-AI-Labs…

Round 5, 24/39: CLBM. Chainlink now plus the last few seconds of Binance mid change gives an estimate; buy the side of the price-to-beat it lands on and hold to settlement. Live-ops tell it's dead: right side, no fill in the final ~2s. Research only. github.com/Polytutor-AI-Labs…

Round 5, 23/39: CBX, the Coinbase cross-PTB scalp. Coinbase mid re-crosses the price-to-beat late in a 5-min window: buy that side, sell into the repricing. Live-ops tell it's dead: the token has already moved when your order lands. Research only. github.com/Polytutor-AI-Labs…

Round 5, 22/39: CBR, the Coinbase range-cross hold. When Coinbase mid crosses the price-to-beat with enough window range behind it, buy once and hold to settlement. Live-ops tell it's dead: confirmed crosses keep flipping before expiry. Research only. github.com/Polytutor-AI-Labs…

Round 5, 21/39: CBP, the Coinbase cross-PTB hold. When Coinbase mid crosses the price-to-beat, rest a limit buy on that side and hold to settlement. Live-ops tell it's dead: fills land after the cross flips back, so you hold losers. Research only. github.com/Polytutor-AI-Labs…

Round 5, 20/39: CBCLP, the Coinbase-Chainlink pair. A wide spread picks the first leg, then you buy the opposite outcome to finish the pair. Live-ops tell it's dead: leg 2 keeps costing too much or never fills, leaving you directional. Research only. github.com/Polytutor-AI-Labs…

Round 5, 19/39: CBCL, the Coinbase-Chainlink spread. When the gap between the feeds gets unusually wide, buy the Up/Down token and sell as it narrows. Live-ops tell it's dead: the spread keeps widening, or the book moved before your fill. Research only. github.com/Polytutor-AI-Labs…

Round 5, 18/39: Binary momentum. Score MACD, RSI and VWAP on 5-min token bars, then check the fill is tradeable. It fails because the indicators are correlated, a score isn't a probability, and by the time they agree the price may be gone. Research only. github.com/Polytutor-AI-Labs…

Round 5, 17/39: Jump-diffusion fair value. Fit a Kou model to BTC returns, turn it into P(S_T > K) for 5/15-min Up/Down, compare to the ask. Live-ops tell: calibration drifts when jump params shift or Binance parts from the settle feed. Research only. github.com/Polytutor-AI-Labs…

Round 5, 16/39: Count-bracket pricing. Simulate a count into exclusive ranges and compare each model probability to the ask. Live-ops tell: a +EV mid dies at the ask; coupled brackets stack risk. Source experiment found no edge after costs. Research only. github.com/Polytutor-AI-Labs…

Round 5, 15/39: Dump & Hedge. After a sharp dump, buy that side, then try to complete the YES/NO pair under $1. Live-ops tell: the complementary book never cheapens, or fills come back unequal, leaving you directional. Research only. github.com/Polytutor-AI-Labs…

Round 5, 14/39: Cross-Venue Arbitrage. Pair related Polymarket and Kalshi contracts so two legs cover every outcome for under $1. Live-ops tell: one leg fills, the other reprices, and you hold a naked position. Check equivalence first. Research only. github.com/Polytutor-AI-Labs…

Round 5, 13/39: Cross-Market Lead-Lag. Use one market as the signal, trade a related one when their usual relationship slips. Live-ops tell: the target reprices before your fill, then both legs lose together. Correlation isn't a contract. Research only. github.com/Polytutor-AI-Labs…

Round 5, 12/39: Fibonacci Retracement & Extension. Anchor an early swing in a 5-min binary token; retracement levels become staged entry zones. Live-ops tell: price keeps walking through your levels into expiry. Levels aren't support. Research only. github.com/Polytutor-AI-Labs…

Round 5, 11/39: Leader-Follower Copy Trading. Mirror a leader's on-chain Polymarket fill in the same token, later. Live-ops tell: your fill keeps landing after the book moved past any leftover edge. Leader PnL isn't follower PnL. Research only. github.com/Polytutor-AI-Labs…