Focal: Signal Credibility Index for Prediction Markets
Maksym Nechepurenko · 2026 · Working Paper
← Research
Quantitative strategies
We study alpha generation, market making, arbitrage, and execution on the hybrid CLOB structure of decentralized prediction markets — venues with unusual liquidity dynamics and binary payoff structures.
What we ask
- —What order-flow signals predict short-term price moves in binary CLOB markets?
- —How should a market maker set spreads when the payoff is bounded at 0 and 1?
- —Where do persistent arbitrage opportunities arise between Polymarket and external odds sources?
- —How does slippage scale with order size in thin prediction market books?
How we approach it
- —Signal construction from microstructure features (ILS, Focal SCI, order imbalance)
- —Market-making model adapted for binary-outcome payoff constraints
- —Arbitrage detection across Polymarket, Kalshi, and prediction-adjacent sports betting markets
- —Execution modeling with empirical slippage curves from historical trade data
Prediction markets are small, illiquid, and bounded — three properties that change almost every classical quant strategy in non-trivial ways. A YES share that trades at 0.80 cannot go above 1.00. Market impact at the tails is asymmetric. The book structure has gaps that would be unexploitable on a deep equity exchange.
These constraints create a research space that is both technically distinct and commercially underexplored. We approach it with the same rigour applied to equity microstructure, translated for the specificities of binary CLOBs.
Publications in this track
Preprint