Independent Probability Estimates for Polymarket BTC Binary Contracts

Summary

This document outlines a research prototype for estimating the terminal direction of BTC in Polymarket 15-minute Up/Down contracts. It combines Binance trade flow and Polymarket reference feeds with the official price-to-beat, time remaining, and forecast volatility to estimate a probability independently of the order book. That estimate is compared with the market’s ask price to identify potential expected value. Entry filters include a fractional Kelly rule, an expected-value threshold, checks on both sides of the book, and estimated round-trip loss. During a position, the prototype monitors changes in estimated value and fixed price exit levels.

The document explicitly says the system is incomplete and supplies no evidence of profitability. The active probability calculation uses a lognormal approximation and empirical volatility windows, but these and many constants lack historical calibration. A separate 43-feature logistic model is disabled for live probability, as are online calibration and the evolution framework. There are no systematic out-of-sample or forward-test results, and execution, settlement, liquidity, manipulation, and feed-latency risks remain insufficiently validated.

Key ideas

• The prototype compares an order-book-independent terminal probability estimate with contract ask prices.
• Its active probability method uses a lognormal approximation and empirical volatility estimates.
• Kelly sizing, expected-value thresholds, order-book checks, and loss limits gate entries.
• The separate 43-feature logistic model and online adaptation systems are disabled.
• No systematic out-of-sample results establish profitability, and execution and market risks remain unresolved.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.