A low max drawdown can hide a short record, low exposure, slow recovery, or a poor return. Ranking 8 ten-year backtests shows what to check besides depth.
Strategy Templates · Updated 2026-10-08
Maximum drawdown is the largest fall from a peak to a later trough in a strategy's equity, expressed as a percentage. It answers a question return alone cannot: how bad did it get on the way? That matters because losses compound against you. A 34.97% drawdown needs a 53.8% gain to get back to the old high (34.97 ÷ 65.03), while a 48.62% drawdown needs a 94.6% gain (48.62 ÷ 51.38). Sorting by drawdown is a sensible first filter. It is a poor last one.
The first is observation length. Maximum drawdown can only record losses that happened during the record. A bot that has run for three weeks in a quiet market has not met a crash yet, so its drawdown is small by construction. Longer records almost always show deeper drawdowns, simply because they have seen more.
The second is leverage. Leverage scales both gains and losses, and it can turn a moderate move into a liquidation. A low drawdown measured in a calm period says nothing about what the same leverage does in a volatile one.
The third is time in market. A strategy that sits in cash or defensive assets most of the time will show a shallow drawdown, but it also misses most of the gains. Low drawdown bought by low exposure is not the same as low drawdown achieved while staying invested.
The fourth is recovery time. Depth is only half of a drawdown. A strategy that falls 15% and recovers in three months is a very different experience from one that falls 15% and takes four years to recover.
The site's eight strategy backtests are rule-based US ETF and stock strategies, re-run on daily data covering roughly October 2016 to October 2026 (runs dated 7 October 2026). Ranked from shallowest maximum drawdown: Vigilant Leveraged Trend Rotation −18.6%, Canary-Signaled Tactical ETF −19.96%, Dynamic Factor Strength −24.07%, Tactical Leveraged Trend Rotation −25.43%, Velocity Strength Filter −27.45%, Fifty-Two Week High Leaderboard −34.49%, Growth-Value Style Selection −34.97%, and Kaufman Efficiency Index −48.62%.
Now divide each strategy's annual growth rate (CAGR) by the size of its maximum drawdown, a Calmar-style ratio of return per unit of pain. Vigilant Leveraged Trend Rotation, first by drawdown, drops to sixth: 10.72 ÷ 18.60 = 0.58. Growth-Value Style Selection, seventh by drawdown, rises to first: 29.54 ÷ 34.97 = 0.84. The full order by this ratio is Growth-Value Style Selection 0.84, Dynamic Factor Strength 0.75, Canary-Signaled Tactical ETF 0.70, Tactical Leveraged Trend Rotation 0.70, Velocity Strength Filter 0.65, Vigilant Leveraged Trend Rotation 0.58, Kaufman Efficiency Index 0.52, and Fifty-Two Week High Leaderboard 0.47.
Vigilant Leveraged Trend Rotation shows the time-in-market problem. Its total return was 175.6% against SPY's 324.0% over the same window. On days SPY rose it captured about 55% of the gain, and on days SPY fell it took about 52% of the loss, the profile of a strategy that spends much of its time out of equities. Its worst full year was 2022, at −10.6% while SPY fell 18.2%. It also shows the recovery problem: its −18.6% drawdown began from a peak on 3 September 2021 and did not regain that high until 12 June 2024, 1,013 days later.
Growth-Value Style Selection, a factor strategy, had the best ratio. Growth-Value Style Selection is an aggressive-tier strategy: it can hold 3x leveraged ETFs (SPXL, TQQQ, UPRO), it was selected from many backtested candidates so luck cannot be ruled out, and it has no live or forward record yet. Its deepest drawdown ran from a peak on 26 January 2018 to a trough on 24 April 2018, −35.0%, while SPY returned −7.9% over the same stretch. It regained the high on 8 January 2020, 712 days after the peak.
The single number hides how similar its next drawdowns were. The second, −29.9%, began on 29 December 2021 and recovered on 13 June 2023, 531 days in all. The third, −29.6%, began on 29 October 2025. Three drawdowns of similar depth describe the strategy better than its single worst one.
Other views add context. 66% of its months ended positive. On days SPY rose it captured about 162% of the index's gain, and on days SPY fell it took about 156% of the loss: the profile of a strategy that can hold leveraged funds and amplifies the market in both directions. A drawdown figure on its own does not show that; the capture ratios do.
The Live Paper Trading record ranks paper-traded crypto bot templates, and some show remarkably small drawdowns. In the delayed snapshot as of 7 October 2026, a Regime Switch Bot on ETH/USDT showed a maximum drawdown of −0.017% and a total return of −0.013% for 9 September to 6 October 2026, 27 days. Its drawdown sub-score is 100 + 4 × (−0.017) = 99.93 and its stability sub-score is 100 − 5 × 0.017 = 99.92, both near the maximum, which helps it to third place with a league score of about 77.2 even though it lost money. Some experimental variants in the same snapshot show drawdowns such as −0.06% on a record covering a single day, 5 October 2026.
A 27-day or one-day paper record and a ten-year backtest are not comparable. The backtest has passed through the 2020 crash and the 2022 bear market; the paper bot may have barely traded. A tiny drawdown over a few weeks is the absence of evidence, not evidence of safety.
It does not tell you the next drawdown will be smaller than the last; the worst one usually lies ahead of any finite record. It does not show leverage, liquidation risk or fees unless those are modelled. It does not separate skill from simply being out of the market. And for a backtest it does not include real execution: these are rules re-run on past prices.
Use maximum drawdown as one column among several. Read it next to the length of the record, the return earned for it, the recovery time, the worst year and month, and how much of the time the strategy was invested. Size any position so that a drawdown deeper than anything in the record would still be survivable. This is education, not investment advice.
Educational content, not investment advice. Guide content and league rankings are independent of any exchange or sponsor relationship.