We measured 22 candlestick patterns on four markets. None beat an ordinary candle.
Hammers, engulfing candles, dojis and morning stars on the daily charts of Bitcoin, Ethereum, Solana and the S&P 500: what happened in the five candles after each one, compared with an ordinary candle.
Every trading course teaches candlestick patterns. A hammer after a decline means buyers are stepping in; a bearish engulfing candle at the top means the move is over. We wanted to know what those patterns actually did on real data, so we measured them.
Short answer: on the daily charts of BTC, ETH, SOL and the S&P 500 (SPY), none of 22 classic patterns showed a statistically clear edge over an ordinary candle in our 27 September 2026 measurement. 30 of the 88 pattern–market combinations were measured and showed no edge; the other 58 appeared fewer than 20 times, too rarely to judge.
How we measured
For every pattern we found each occurrence with a deterministic detector — the same one you use in Evident, with fixed definitions for body size, shadows and the preceding trend. Then we asked one question: in the five candles after the pattern, did price move in the textbook direction more than it does after an ordinary candle?
The comparison with an ordinary candle matters more than anything else. Bitcoin went up over most of the period, so almost any "bullish" signal looks good if you only count how often price rose afterwards. What matters is the difference from the baseline, and whether that difference is large relative to its noise (the t-statistic).
A pattern gets a verdict only if it appeared at least 20 times on that market:
- works — a clear edge in the textbook direction
- inverse — a clear edge in the opposite direction
- no edge — measured, and indistinguishable from an ordinary candle
- too few to judge — fewer than 20 occurrences
What we found
| Pattern on BTC, daily | Occurrences | 5-candle edge vs ordinary | t |
|---|---|---|---|
| Spinning top | 214 | −0.23% | −0.80 |
| Doji | 172 | −0.43% | −1.71 |
| Bearish engulfing | 72 | +0.63% | 1.09 |
| Bullish engulfing | 70 | −0.19% | −0.27 |
Two things stand out. First, the most famous reversal pattern of all points the wrong way on Bitcoin: after a bearish engulfing candle, the next five days were on average slightly better than after an ordinary day, and after a bullish one slightly worse. Second, neither difference is anywhere near significant. With a t-statistic around 1, you cannot tell the pattern apart from luck.
The rarer patterns — morning and evening stars, three white soldiers, tweezer tops — simply did not happen often enough on daily candles to say anything. That is also a finding: a pattern you see a dozen times in four years cannot be the foundation of a strategy.
What this does not mean
It does not mean candles are useless. A pattern can still help as a filter inside a rule — for example "hammer and price above the 200-day average and volume above normal". Context can change everything, and that is exactly what you can test in Evident: every pattern is available as a condition in the strategy editor and as cdl("hammer") in the indicator lab.
It also does not cover other timeframes or markets. The measurement is for four markets on daily candles. On your market and your timeframe the answer may differ — so measure it.
Check it yourself
The live study is recomputed every day from market data and shows every pattern on every market, including the ones that were too rare. In Markets → Patterns you can run the same measurement on any symbol and timeframe in a few seconds.
This is research, not investment advice. Past behaviour of a pattern says nothing certain about the future.
Evident does not give investment advice. Past performance does not indicate future returns. Leveraged and automated trading carries the risk of losing capital.