Grading track records by drawdown, expected value, and ruin

An editorial reading order

TradersWeek editorial reading: treat a performance table as a survival document. First ask whether peak-to-peak loss stayed inside a drawdown limit. Then ask whether leftover months still show positive expected value. Only then ask whether the path between large up months is still a livable risk-of-ruin problem.

The filters themselves come from the archive workflow. The three-question order is editorial. It is not attributed to the archive as a required sequence, and it is not a present-day grade of any account.

Start with the drawdown limit

Control of peak-to-peak equity drawdown across an entire multi-year sample is treated as a drawdown-limit screen for whether risk stayed contained, not merely whether a gain appeared. On a track record the drawdown limit is that same bound applied to peak-to-peak equity loss, meaning the largest decline in equity from a prior high to the next low inside a stated evaluation window.

A first-look evaluation filter is the same-period ratio of annual net gain to worst peak-to-peak drawdown. Many records cluster near two-to-one. Assuming every program shares that ratio is described as unsound.

A stricter drawdown-limit bar is a single-digit peak-to-peak loss with a same-period gain three to five times that loss, repeated across three successive calendar years.

Then read leftover expected value

Expected value asks whether leftover outcomes still pay after the spectacular months and unmarked open profit are stripped out.

An expected-value reading of unmarked profit treats about 18% to 20% conversion of aggregate open profit into closed profit as typical and about 25% as a higher bar, so most open profit is not expected to become spendable equity.

Closed-equity change is preferred over open-equity change for evaluation because open profit cannot be spent and a large share of it is not expected to be realized. Closed equity is the marked balance that has already been realized and can be spent. Open equity cannot be withdrawn.

Theoretical cost inside rolling windows

Theoretical cost, used as an expected-value filter, drops every double-digit up month inside a twelve-month window and sums remaining single-digit gains with all losses to show the path cost of reaching the large months.

A nine-year monthly series contains 97 overlapping twelve-month windows, so one January-through-December year is only one evaluation interval among many. A rolling twelve-month window is every consecutive twelve-month slice inside a longer monthly series, not only the calendar year.

In a ten-month illustration, removing three double-digit up months left a 33% residual loss while the full year finished down 4%, showing a calendar result can hide a large cost of reaching outlier months.

Then ask about ruin on the leftover path

Risk of ruin asks whether the path between large profitable months still leaves the account intact.

If leftover months after large up periods are removed stay positive, those months are a bonus. If they are a large loss, the path is framed as a risk-of-ruin problem because vulnerability in non-profit stretches moves the account closer to failure.