Chart patterns and candlestick setups
Does buying a moving-average reclaim pullback work?
Rejected Pre-registered and tested, July 2026
A positive average, a negative median, and one year of 2013 carrying the entire result.
What the pattern claims
Price closes back above its 20-day exponential moving average on at least 1.5 times normal volume, with the shorter average above a rising longer one. The next day price pulls back but holds above the moving average. You buy that hold. The claim is that the reclaim signals the trend resuming and the pullback gives you the entry.
How we tested it
S&P 500 companies, tested across four variants, with companies that later delisted included.
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
The base signal returned +1.09% on average at 60 days — the only positive headline number in this family. But the median case lost 1.01%, only 45.6% of trades were profitable, and the second era was negative. Short horizons were flat or negative throughout.
A single year, 2013, carried the entire average.
What it means
We then did something that turned a puzzling result into a clear one. We capped how far the reclaim candle could stretch, on the reasoning that if the pattern works as described, removing the most extreme examples should barely matter.
It made the result worse, and pushed the cell negative. Which proves the positive average was never coming from the pattern at all — it was coming from a handful of extended momentum spikes that happened to satisfy the definition. The thing being tested was not the thing producing the number.
One narrow survivor is worth naming honestly. Requiring MACD to be sloping upward on the confirmation bar produced a genuine short bounce: +0.41% average over three days, 54.8% of trades profitable. But three-day trades cost roughly 0.40% per round trip to execute, so the edge is approximately equal to the cost of capturing it, on 115 examples. That is not a strategy. It is a measurement of the fee.
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 buying the dip after a spike work? The first pattern in this family to lose money significantly rather than merely do nothing.
- 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.