Encyclopedia › Chart and candlestick patterns ›
Head and shoulders
Not tested by us Not tested by us yet. No verdict stamp.
Strength: mixed. Lo, Mamaysky and Wang (2000) looked at US stocks from 1962 to 1996. They found that several patterns, for example head-and-shoulders and double bottoms, "provide incremental information and may have some practical value". They give no success-rate percentages, and they did not test profits after costs. Savin, Weller and Zvingelis (2007) found little or no support for trading it on its own.
Picture a person: left shoulder (a peak), head (a higher peak), right shoulder (a lower peak). The line under the two dips is the 'neckline'. Upside down, it is an inverse head and shouldersThree peaks, the middle one highest; often seen before a trend turns..
What traders claim: a break below the neckline means the uptrend is over.
Savin, Weller and Zvingelis (2007) studied head-and-shoulders on US stock indices (1990–1999). They found 'little or no support' for trading it on its own. But the pattern did help predict returns when combined with holding the market. Lo, Mamaysky and Wang (2000) also included it in their detector.
Evidence: see Do chart patterns work? What the research says. In short: mixed, and very sensitive to how you define the pattern.
Related
Sources & studies
- Investopedia (secondary): Head and shoulders pattern
- Savin, Weller & Zvingelis (2007), Journal of Financial Econometrics
- Lo, Mamaysky & Wang (2000), Journal of Finance (author PDF)
Last checked: 2026-11-04 · Educational, not financial advice.
Education only, not financial advice. No signals. Backtests are hypothetical: real trading can be worse. Past results don't predict future results.