Bitcoin Cross-Exchange Statistical Arbitrage and Trading-System Lessons
Summary
This retrospective describes an early Bitcoin trading effort that began with cross-exchange price gaps and evolved into a statistical arbitrage strategy. Directly buying on one venue and transferring coins to another exposed traders to transfer delays and price changes. The alternative kept positions on both exchanges, trading when their spread moved away from its historical average and reversing when it moved back. The approach depended on the spread remaining sufficiently stable, and the account explains how exchange access, currency conversion, liquidity, and operational constraints shaped the opportunity.
The authors also describe building data collection, signal, and order systems, including browser automation where exchange APIs were unavailable. Unequal fills required sequential orders or follow-up trades to control residual exposure. The authors report strong performance over their operating period, but these are retrospective self-reported figures, not independent validation. The strategy was eventually closed as volumes and spreads declined, work commitments changed, and concerns about the safety of exchange funding increased. Exchange failure and changing market structure remain central limitations.
Key ideas
- Cross-exchange price gaps can be traded without transferring coins during each opportunity by holding inventory on both venues.
- The described statistical arbitrage enters when the spread departs from its typical level and exits as the spread converges.
- Transfer delays, venue access, fiat conversion, and exchange reliability can determine whether an apparent price gap is tradable.
- Unequal execution across venues creates residual exposure that requires follow-up orders or carefully sequenced execution.
- The reported historical performance is anecdotal and depended on market conditions that later changed.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.