Begin with the composition of the drawdown
Measure how many trades contributed, whether one or two extreme losses dominate, and whether the drawdown arrived gradually or in a tight cluster. A ten-trade losing streak and one abnormal execution loss are different research problems.
Segment the period
Compare the drawdown period with profitable periods by direction, session, holding time, volatility and market regime. Changes in the mix of market conditions can make a stable strategy appear to have "stopped working" even when the underlying relationship is behaving as expected.
Check trade-path deterioration
Compare MAE and MFE before and during the drawdown. If MFE collapses, entries may be receiving less follow-through. If MFE stays healthy but realized results deteriorate, exit efficiency or giveback may be the larger issue.
Watch for concentration risk
Some systems rely on a small set of outlier winners. A drawdown can emerge simply because those rare trades have not occurred recently. Analyze the distribution of contribution rather than relying only on average trade statistics.
Do not confuse adaptation with curve fitting
A drawdown is emotionally compelling evidence, but not necessarily statistically sufficient evidence for a strategy change. Any proposed mitigation should be tested against prior periods and ideally against untouched data.
Ask better questions about your strategy.
EdgeQuery is being built to analyze automated trading-system history beyond headline statistics. The founding beta focuses on trade-level behavior, market context, failure patterns and research priorities.