Do trading strategies reduce drawdown versus buy & hold? Their worst falls were shallower at every reported percentile — and most still lagged on return.
Across the famous US stocks & ETFs in our frozen census, the distribution of strategy drawdowns sat shallower than simply buying and holding at every reported percentile — yet the same backtests trailed buy & hold on return 86% of the time.
A shallower drawdown bought with a lower return is avoidance, not skill — and not an edge.
As of September 25, 2026 · frozen census release · DOI 10.5281/zenodo.22974116
This is a frozen, reproducible snapshot — not a live number and not a promise about the future. It reads a survivorship-biased set of well-known US stocks & ETFs over one roughly five-year window, each against its own buy & hold, net of modelled costs, with no out-of-sample split, and it is not significance-tested. It compares two distributions, not a per-asset paired reduction.
The finding, both halves
At every reported percentile, the strategy drawdowns came in shallower than holding. But the shallower fall was not won as an edge: the very same backtests — 721 of 839 judged stock & ETF backtests (86% of them, only 14% beat) — still lagged buy & hold on return. The smaller hole was paid for in foregone return, not earned.
839 judged backtests over 36 famous US stocks & ETFs and 25 strategies. These are two distributions set side by side, not a per-asset head-to-head. Each column is read across the same judged backtests — every famous asset carries its own buy & hold, counted once for each strategy tested on it — so the honest unit to keep in mind is the handful of underlying assets, not the raw row count. There is no significance test here, and the single worst and mildest rows are each one extreme observation — and because every asset's buy & hold is repeated once per strategy tested on it, the buy & hold extremes rest on far fewer distinct assets — so lean on the central band, not the extremes.
Maximum drawdown by percentile — strategy vs buy & hold (US stocks & ETFs)
| Case | Strategy | Buy & hold |
|---|---|---|
| Worst case seen | −87.9% | −90.9% |
| A deep fall (worst ~10%) | −52.7% | −74.9% |
| A deeper-than-typical fall | −40.3% | −55.5% |
| Typical (median) | −27.4% | −35.1% |
| A milder-than-typical fall | −18.6% | −25.9% |
| A mild fall (mildest ~10%) | −13.9% | −20.8% |
| Mildest case seen | −5.4% | −17.3% |
Read down the middle of the table (the typical and near-typical rows), not the single worst/mildest extremes. Both columns are losses — a shallower number is not a better outcome on its own, only a smaller fall.
The gap was widest in the deepest falls (about 22.1 points apart in the worst tenth) and narrowest among the mildest (about 6.9 points) — the pattern time out of the market would produce: sitting out part of a decline spares the most exactly when the decline is worst, and barely anything when there was little to avoid.
Why — avoidance, not skill
The likeliest reason is simply time out of the market: a rule that steps aside through part of a decline avoids part of the fall — but it also misses part of the rebound, which is exactly why most still trailed buy & hold. This study didn't measure how much time each rule actually spent in cash, so that is the honest explanation, not a measured cause. A shallower drawdown bought with a lower return is avoidance, not skill — and not an edge.
The other slices, as context only
On crypto the gap looked larger still — median −67.5% vs −85.3% — context only — a small, volatile, survivorship-biased sample; an overfitting artifact, not a demonstrated edge; not significance-tested
Pooled across stocks and crypto the medians were −33.4% vs −45.3%, but that blend is composition-dependent — more crypto in the mix pulls it around — so it is context, not a citable figure. The stocks slice above is the headline.
These drawdowns come from famous names that still trade today. Names that delisted, went bankrupt or (for crypto) died aren't in the set — and those include the deepest holes of all — so the full real-world range of drawdowns is wider than the slice shown here. And these are the strategies' drawdowns, not the assets': an exit-based rule can sidestep part of a fall a buy-and-holder would have sat through.
One caveat cuts the other way for the buy & hold column: the names that delisted or went to zero — the deepest holes a buy-and-holder could ever suffer — aren't in the set, so the real cost of simply holding is understated here, and the gap shown is, if anything, conservative.
Test a rule on the asset you care about — free, no sign-up.
These are simulated backtests over roughly a five-year window, frozen as a citable release — hypothetical results, not live figures and not advice. Past performance is not a reliable indicator of future results. This is an educational, analytical tool — not investment advice, and not a recommendation to buy or sell anything. You make all decisions and execute on your own broker.