Does a median above the average mean a strategy is better?

Not on its own. It means the typical ticker did slightly better than the headline average suggests, which is the opposite of the usual warning on this site, but bollinger squeeze breakout's 0.27 percent median is still a modest result over a full decade. Direction of the gap matters for how to read the headline number. It does not by itself make the underlying result strong.

Why don't a few big winners pull this study's average up the way they do on other strategies?

Because its most extreme results are capped. The best ticker in the completed sample gained 12.36 percent and the worst lost 11.34 percent, both far smaller than the triple- and quadruple-digit outliers that appear in trend-following studies like golden cross or 52 week high momentum. Without an outlier large enough to dominate the calculation, the average return tracks closer to how the bulk of tickers actually did, and here slightly more of them lost money than gained it.

Guide · 6 min read

Bollinger squeeze breakout is the only strategy here where the median beats the mean

Every other strategy study on this site has an average return sitting above its median, pulled up by a handful of large winners. Bollinger squeeze breakout runs the other way: its 0.27 percent median beats its 0.17 percent average, because its worst tickers, not its best, are doing the pulling.

Updated 2026-09-13

Bollinger squeeze breakout's median ticker returned 0.27 percent. Its average ticker returned 0.17 percent. That ten hundredths of a point gap runs backwards from every other strategy study published on this site, where the average sits above the median, not below it.

Tenachine's guide on reading a return distribution explains why the average usually sits higher: a losing trade is bounded by its stop, a winning trade is not, so a handful of large winners drag the mean up past the middle of the pack. Tenachine's guide on triple ma pullback shows a sharper version of that same pattern, where the average stayed positive while the median ticker actually lost money. Bollinger squeeze breakout inverts the direction entirely. Here the median, not the average, is the higher number, which means whatever is pulling this study away from its center is happening on the downside, not the upside.

Bollinger squeeze breakout, 161 completed tickers, by return bucket

Add the three loss buckets and 40.4 percent of tickers lost money, against 38.6 percent that gained. Neither the best ticker, 12.36 percent, nor the worst, negative 11.34 percent, is large enough to dominate the average the way a single outlier does on other studies published here.

Where the reversal comes from

Bollinger squeeze breakout's most extreme results are unusually tame for this site. The best ticker gained 12.36 percent. The worst lost 11.34 percent. Compare that to golden cross's best result, 119.41 percent on a single ticker, or 52 week high momentum's 3300.42 percent NVDA trade, both large enough on their own to lift an entire study's average well above its typical result. Bollinger squeeze breakout never produces a winner in that range, so there is no single ticker doing the average's work. Instead, 40.4 percent of tickers lost money against 38.6 percent that gained, a narrow tilt toward the loss side that is just wide enough, combined with capped upside, to pull the mean below the median rather than above it.

Coverage is a separate issue worth naming rather than folding into this one. Tenachine's guide on survivorship bias already flags this study's 59 failed runs out of 220, a 26.8 percent failure rate, as one of the worse coverage gaps on the site. That gap affects how much of the universe these numbers describe. It does not explain why the mean and median sit in this order relative to each other among the 161 tickers that did complete a run, which is a separate question about the shape of a result, not the size of the sample behind it.

What this changes about reading the headline number

Every other mean-versus-median comparison on this site is a warning that the average return overstates what a typical ticker actually did. This one runs the other way. For a reader used to that warning, checking the median on this particular study would have been reassuring rather than cautionary, since it comes in higher than the 0.17 percent headline, not lower. Both directions are useful to know before repeating either number, and neither direction is something a reader could have guessed from the headline alone.

None of this makes bollinger squeeze breakout a better or worse strategy than any other studied here. A 0.17 to 0.27 percent range over a decade is a modest result either way, closer to noise than to a strong edge on its own. Past performance does not predict future results, and a strategy whose mean and median happen to sit in an unusual order this decade is not guaranteed to keep that order in the next one.