Backtested Scanners
Crash Reversal
71 Hits · last calculated September 23, 2026 Source: price data (daily closing prices) · Market filter active: the list shows 0 hits from Germany
Methodology & criteria
The buy day AFTER the crash, in the form we back-tested for you. Four stages have to occur in order: a death cross (the 10-day line falls below the 20-day line), then a drop of at least 20 percent from the close on the cross day, then capitulation — at least one daily close 15 percent or more below the stock's own 20-day line — and finally the first daily close back above the 5-day line. That day is the signal. If the price recovers above its 20-day line before the 20 percent mark is reached, the case does not count. The measured rule includes the exit too: sell at the first close 11 percent or more above the entry price, at the latest at the close of the 12th trading day. Across 9,908 trades in 5,008 US stocks since 2010, survivorship-free and after trading costs: median +6.4 percent per trade, hit rate 66.6 percent, 1.87 percentage points ahead of the S&P 500 (robust mean). The other side comes from the same calculation and is shown in the hit list: on crashes of 50 percent or more this short recipe loses against the index (−5.4 percentage points across 1,116 cases), and holding long is no reliable alternative there — a pre-registered follow-up test does show large edges, but strip out the 2020 and 2021 entry years and all four tested long-hold variants fall behind the index. That is why the "Crash Depth" column states how far each hit fell, and a badge marks the cases at 50 percent or more. The stock has to stay tradable: at least $1 unadjusted and at least $1 million of median daily turnover. A hit stays on the list for twelve trading days — as long as the measured trade would still be open. Source: price data.
No global trading filters — this strategy checks the entire stock universe purely against its own criteria (mega caps over $50B included).
🔔 Watch scanner
Hit List
Tip: clicking a column header sorts the table by that column; a second click flips the direction.
| Symbol | Signal Age (Days) | Crash Depth | Earnings | Avg/Y 3Y | Volatility | Stage | Funda Rating | Piotroski | MktCap | Industry | AI Rating | Deep Dive | Deep-Dive Report | Sector | Price | YTD | 6 Mo. | 1 Year | Off High | Price Target | RS | EPS Rating | ADR 10D | ADR 30D | Beta | P/E | P/E (f) | P/S | P/B | P/FCF | PEG | EV/EBITDA | EBIT Margin | Gross Margin | Net Margin | ROE | ROA | Debt/Eq | Equity Ratio | Sales +/Y | Growth Score | Div. Yield | Payout Ratio | Altman Z | Inst. % | Short % | Analysts |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No stocks currently pass this scanner. | |||||||||||||||||||||||||||||||||||||||||||||||
Frequently Asked Questions
The buy day AFTER the crash, in the form we back-tested for you. Four stages have to occur in order: a death cross (the 10-day line falls below the 20-day line), then a drop of at least 20 percent from the close on the cross day, then capitulation — at least one daily close 15 percent or more below the stock's own 20-day line — and finally the first daily close back above the 5-day line. That day is the signal.
All scanners are recalculated daily across the entire stock universe — most recently on September 23, 2026. The data basis is fundamental data and SEC filings (10-K annual reports and 10-Q quarterly reports).
Currently, 71 stocks pass this scanner's criteria (as of September 23, 2026).
No global trading filters — this strategy checks the entire stock universe purely against its own criteria (mega caps over $50B included).
The scanner looks for the buy day AFTER a crash. Four stages have to occur in order. First, a death cross: the 10-day line falls below the 20-day line, on a daily-close basis. Second, a real crash: from the close on the cross day to the lowest subsequent close, the stock loses at least 20%. If it recovers above its 20-day line before that, the case does not count — that was a dip, not a capitulation. Third, the capitulation itself: at least one day closes 15% or more below its own 20-day line. And fourth, the buy: the first daily close back above the 5-day line, no earlier than the capitulation day and after the 20% mark has been reached. That day is the signal. The stock also has to stay tradable: at least $1 unadjusted on the cross day and at the low, and at least $1 million of median daily turnover from 30 trading days before the cross through the buy. The measured rule also includes an exit no scanner can enforce: sell at the first close 11% or more above the entry price, at the latest at the close of the 12th trading day. We back-tested this rule across 9,908 trades in 5,008 US stocks since 2010, survivorship-free — delisted names stay in the data set. Under a realistic cost model (0.5%, 0.2%, or 0.1% per side by trading liquidity, while the benchmark deliberately carries no costs at all): median +6.39% per trade, robust mean +3.70%, hit rate 66.6%, an edge over the S&P 500 of +3.47 percentage points on the median and +1.87 on the robust mean. Two things belong to that honestly. First, the edge only comes from an early entry AND a short exit combined — buy only above the 20-day line and the rule beats the index — measured on the winsorized mean — in none of 1,500 tested exit variants, and stop-losses worsen the result across all 50 tested distances. Second, on crashes of 50% or more the picture reverses: 1,116 such cases came in 5.44 percentage points behind the index. Holding long is no reliable alternative there: a pre-registered follow-up test does show large edges, but strip out the 2020 and 2021 entry years and all four tested long-hold variants fall behind the index. The "Crash Depth" column and the "Crash of 50% or More" badge show exactly that on every hit. All figures and caveats are in the study. A hit is a find, not a buy signal. Source: price data.
Related scanners
Note: pure fact-based analysis, not investment advice and not a solicitation to buy or sell. All figures without guarantee.