Using the Cold Blood Index to Assess Trading Drawdowns

Summary

The article introduces the Cold Blood Index (CBI) as a way to judge whether a live trading strategy’s drawdown is consistent with losses that could have occurred in its backtest. It compares the observed drawdown depth and duration with historical windows from the strategy’s balance curve. The fraction of comparable windows that reached at least the observed loss provides an estimate of how unsurprising that drawdown is: a higher value suggests the event is more consistent with historical behavior, while a lower value can indicate a possible departure from the backtest.

The discussion covers both a strategy that is underwater from inception and a drawdown that begins after live trading has started. It also presents a resampled version that bootstraps the historical balance curve and reports percentiles across repeated samples, reflecting variation in trade outcomes. This is a diagnostic aid, not an automatic retirement rule. Its conclusions depend on the representativeness of the backtest, and resampling can be misleading when returns have meaningful serial correlation because it disrupts that dependence.

Key ideas

• The Cold Blood Index compares a live drawdown with historical drawdowns of similar depth and duration.
• A higher index suggests the observed loss is more consistent with the backtest’s historical experience.
• The method can account for drawdown duration relative to the time elapsed since live trading began.
• Resampling gives a range of index estimates but can distort strategies with serially correlated returns.
• The index is a diagnostic for strategy monitoring, not a standalone decision to stop trading.

Tags

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