Recomputing a statistic over a moving window to see how it evolved, rather than collapsing the whole history into one number.
A single Sharpe ratio for a ten-year record hides everything. The same figure computed over rolling twelve-month windows shows whether it was 1.4 throughout or 3.0 for two years and 0.2 for eight.
Choose the window by trade count, not calendar convenience: it must be long enough that the statistic is not noise - see sample-size-for-edge - and short enough to reveal change. For a system trading 300 times a year, a rolling 150-trade window is a reasonable compromise.
Read the trend rather than the last point. A rolling expectancy that has declined steadily across three windows is a different signal from one that dropped in the most recent one, and only the first justifies acting. The last point of any rolling series is also the noisiest.
Original diagrams for the ideas on this page. Illustrative, not real market data.
Expectancy: the average trade. Forty trades sorted by outcome: 24 small losses and 16 larger wins. Weighting each side by how often it happens gives the average result per trade, marked here by the dashed line at +$120.
Educational only, not advice. Spotted an error? Post in Site Feedback.