Underwater period (time to recover)

An underwater period is the stretch of time during which an equity curve stays below its previous peak, from the trade that set the high to the trade that makes a new one. Where maximum drawdown measures how deep the curve fell, the underwater period measures how long it took to recover.

Senzoukria · Glossary · Updated September 2026


Depth versus duration

Most performance reports quote a maximum drawdown: the largest drop from an equity peak to the trough that followed. That number says how much was lost, but nothing about how long the account sat there. Two strategies can share the same maximum drawdown while one recovers in a week and the other takes half a year. The underwater period is the second dimension: the elapsed time, in trades or in calendar days, between an equity high and the next equity high that exceeds it.

A stretch that has not been recovered by the end of the record is called open. It has a start but no end yet, and its duration is only a lower bound. Reporting it as recovered would overstate the strategy.

  • Start of a stretch: the trade that sets a new equity high.
  • End of a stretch: the first later trade whose cumulative result exceeds that high.
  • Longest underwater period: the maximum of all stretch durations, open stretches included with their partial length.

Why the duration matters more than it looks

A trader rarely abandons a strategy because of a single loss. Abandonment usually happens during a long flat or declining stretch, when the rule still works but the account has not made a new high for weeks. The underwater period puts a number on that psychological load. It also matters for funded or evaluation accounts, where a time limit or consistency rule can expire before the recovery arrives.

  • Compare the longest stretch with the length of the whole sample: a stretch that covers most of the record means the curve is dominated by one recovery.
  • Count the stretches, not only the longest one. Many medium stretches describe a choppy curve; one long stretch describes a regime that stopped working for a while.
  • Measure the same way in the backtest and in the journal, otherwise the two are not comparable.

In Senzoukria

The backtest results panel in the desktop includes a chart named "Longest stretches under water". It lists each stretch with its length in days and marks it as "recovered" or "still open"; its hint states that maximum drawdown gives the depth while this chart gives the time without a new equity high. When every trade made a new high, the panel says so instead of drawing an empty chart.

The journal's "Previous sessions" view draws the session equity curve with an "Underwater" series beneath the "Equity" series. Both are computed on closed trades: the intra-trade dip would require replaying the ticks, and the panel notes that it does not claim to.

Common mistakes

  • Reading an open stretch as if it had ended: its duration is a minimum, not a result.
  • Quoting maximum drawdown alone and assuming recovery was quick.
  • Measuring the stretch on a curve without transaction costs; costs lengthen every recovery.
  • Mixing closed-trade equity and mark-to-market equity in the same comparison; the second dips deeper and longer.

This page in other languages

Frequently asked questions

Is the underwater period the same as the drawdown duration?
They describe the same interval. Drawdown duration usually refers to a single episode from peak to recovery, while an underwater analysis lists every episode and singles out the longest. Both start at an equity high and end at the next high that exceeds it.
Can a strategy have a small drawdown and a long underwater period?
Yes. A strategy that loses a little and then trades flat for a long time has a shallow drawdown but a long recovery. That profile is easy to overlook when only depth is reported, and it is often the one that gets abandoned in practice.
Why does Senzoukria compute the underwater period on closed trades only?
Because the equity path inside a trade is not known unless the ticks of that trade are replayed. The journal and the backtest panel state this limitation rather than estimating the intra-trade dip. A mark-to-market curve would show deeper and longer stretches.

Keep reading