Chart patterns and candlestick setups
Does buying the dip after a spike work?
Rejected Pre-registered and tested, August 2026
The first pattern in this family to lose money significantly rather than merely do nothing.
What the pattern claims
A story with a mechanism behind it, which is more than most patterns offer. Some catalyst spikes a stock hard. Short sellers attack it and drive it into a retracement. The dip gets absorbed on drying volume — sellers running out rather than buyers giving up — and then a second, more sustained wave follows. The claim was that this works best in small companies, where a single seller moves more.
How we tested it
354 occurrences across 294 companies in the S&P 500, 400 and 600, 2006–2026, measured at 20 days.
Every study here was pre-registered: the exact rule, the pass and fail thresholds and the data window were written down and cryptographically fingerprinted before the test was run. That makes it impossible to move the goalposts after seeing the answer.
Results assume you buy at the next morning's open, not at the closing price that triggered the signal, and they include companies that were later delisted or went bust. Returns are measured against what the rest of the market did on the same days, so a rule that made money only because it fired on days everything rose scores zero here.
What happened
−1.47 percentage points at 20 days against what a typical stock did on the same days, with only 42.9% of trades profitable. Measured against the S&P 500 instead, −1.68 percentage points. The two agree.
That matters because it is the first member of this family to land clearly below zero rather than sitting in the noise. Most chart patterns we test are simply inert. This one was mildly harmful.
What it means
The most useful thing this study produced happened before the test ran at all.
The rule as originally proposed — a 3× volume surge with a 15% close — produced 1,531 surges, which narrowed to 534 dips, which narrowed to just 85 complete signals, of which 39 were small caps. That is far too few to conclude anything from. Relaxing the thresholds to 2× volume and 10% got us to 391, and that relaxation was decided on counts alone, before the test was frozen, without looking at a single return.
The lesson is about multi-stage patterns generally. A three-stage funnel lost 93% of its starting points. If your rule has several conditions that must happen in sequence, count how many complete examples history actually contains before you spend any time on it. Most such rules do not have enough examples to be testable, which also means they do not have enough examples for your own eyes to have learned anything reliable from them.
For the record, the strict version was worse, not better, so nothing was lost by relaxing it.
A rejection is not "this never works"
It means: tested on this universe, over this period, against a bar written down in advance, it did not clear the bar. Where a test lacked the power to decide either way we record it as undecided rather than rejected. The full research ledger has all 192 experiments and what the failures have in common.
Other patterns we tested
- Does the bull flag pattern work? The pole is the only live part of the pattern. Every flag condition made the result worse.
- Does MACD divergence work? The loose version changed the outcome by less than one hundredth of a percent. The strict version widened both tails.
- Does breakout and retest work? The setups the rule throws away outperformed the ones it keeps.
- Does a big RSI drop mark a buying opportunity? The family's only statistically significant result, and it vanished completely once we accounted for when it fired.
- Do three shrinking red candles signal a bottom? The only chart pattern that beat all its own control groups — and it still could not clear the error bars.
- Does a tight base after a sharp drop work? Removing the base conditions improved the result. The base was the problem.