A more patient idea
A squeeze strategy waits for volatility to compress, then takes the expansion when it comes. The premise is appealing because it is selective by construction: no squeeze, no trade. Fewer trades means fewer fees, fewer whipsaws, and less exposure to being wrong.
Of the strategies we graded this week, Keltner squeeze is the one where that premise showed up in the data.
What we tested
Our full backtest window on the top 50 coins by market cap, with fees and slippage applied to every trade (the exact window is shown on the settings page). Then the exit grid: 5 stop-loss levels x 6 take-profit levels = 30 full backtests. The live grid refreshes with our data pipeline, so the exact cells move; the shape does not.
Live grid: Keltner Squeeze — stop-loss & take-profit settings.
Result: 0 of 30 combinations profitable — but the shallowest losses of the four by total return
As of 2026-07-26, the best profit factor is 0.82. A profit factor under 1.00 means the losing trades took more than the winning trades brought in. It is the number to look at first, because unlike a return figure it does not depend on how you sized positions.
The interesting number is the trade count: 2,140 to 2,424 across the whole 50-coin universe, depending on exits. Compare that to 6,000–11,500 for the Donchian breakout and 7,300–16,400 for Heikin-Ashi on the same universe and window. The squeeze filter did what it claims to do — it traded roughly a quarter as often.
And it worked, in the direction you would hope: the squeeze grid’s losses are the shallowest of the four strategies we graded this week measured by total return. On profit factor — the metric we told you to look at first — mean reversion has been level with it or slightly ahead across recent refreshes. Selectivity is not nothing. It just did not cross zero.
Update (2026-08-15): This strategy is now killed
When this post was published (2026-07-26), the Keltner Squeeze carried under test status — the grid losses were shallow enough to warrant continued monitoring. A fresh out-of-sample regime-robustness test run on 2026-06-28 settled the question: the SHORT edge is bear-beta. The strategy’s profits in the OOS window came from a falling market, not from the squeeze filter functioning as designed. Adjusted for that regime exposure, no standalone edge survives.
We said we would publish the kill decision. Here it is. The strategy now appears in the killed column on the research log with the full kill rationale on the strategy page.
The honest takeaway
Being selective made the losses smaller. It did not make them profits. The Keltner squeeze is the most interesting of the four strategies we published verdicts for this week, and it is still not something we would put money behind.
Live table: Keltner Squeeze settings. Everything else, including what we killed, is in the research log.