One strategy's 94 percent return comes with a 57 percent failure rate
The single best headline return across every strategy study on this site belongs to the study with the worst completion rate. Read one number without the other and the 94 percent looks like a discovery instead of a warning.
The best headline return of any strategy study published here is 94.33 percent, and it belongs to 52 week high momentum, a rule that buys a stock after it makes a new 52-week high. The same study also has the worst completion rate of any strategy study published here: 126 of 220 tickers, 57.3 percent of the universe, failed to produce a usable backtest.
The 94 tickers that did complete a run show numbers that look almost too clean to be real: an 89.45 percent win rate and an aggregate risk to reward ratio of 287.51, meaning winning trades outweighed losing trades by close to three hundred to one. Tenachine's guide on survivorship bias already flagged this study's 57.3 percent failure rate as the worst of four compared there. That guide did not show what the surviving numbers looked like. Here they are, and they are the most extreme figures on the whole site.
What a 57 percent failure rate should make you ask
Tenachine does not currently publish why any individual ticker's run failed, so this is a real unknown rather than a solved one. What can be said plainly is that a result built on 94 of 220 tickers, fewer than half the universe, is not entitled to be read with the same confidence as one built on 214. The question worth asking before trusting the 94.33 percent is not whether the math behind it is correct, it almost certainly is, it is whether the 94 tickers that survived are a fair sample of what the rule would do across the full 220, or whether the rule itself only produces a completable backtest on the names that were already the strongest momentum stories in the set. A rule built around chasing new highs failing more than half the time it is tried is not a small footnote to the headline number. It may be most of the explanation for it.
The mirror strategy, without the same gap
Week 52 low reversion, a rule built on the opposite idea, buying a stock after it makes a new 52-week low, ran on the same 220 stocks and completed 214 of them, failing only 6, a 2.7 percent failure rate against 52 week high momentum's 57.3 percent. Its numbers are far more modest and far more believable: a 21.25 percent return and a 64.7 percent win rate. But its average maximum drawdown, negative 27.21 percent, is actually deeper than 52 week high momentum's negative 16.37 percent. The well-covered, believable strategy does not win on every dimension either. It only wins on the one that matters most for deciding whether to trust the headline number at all: how much of the universe the number is actually describing.
A completion rate belongs next to a return the same way a drawdown belongs next to a win rate. The 94.33 percent figure is not fabricated and nothing here says the underlying trades did not happen. It says that a return this far outside every other number on the site, sitting on top of the thinnest sample on the site, is a reason to look harder before repeating it, not a reason to lead with it. Every study page here publishes its completion count next to its return for exactly this reason, and this study is the clearest case for why that pairing matters more than the headline number itself.