Classic Chart Patterns (Head & Shoulders, Cup & Handle, Three Drives)
Visual price formations claimed to predict continuation or reversal.
Mixed and weak evidence — some patterns carry marginal information, but limited tradeable value after costs and identification is subjective.
- Why it fails
- The evidence is mixed and weak: Lo, Mamaysky & Wang (2000) found some patterns carry marginal information but limited tradeable value after costs, and pattern identification is subjective across analysts.
Head and shoulders, cup and handle, three drives — the classic chart patterns are the visual grammar of retail technical analysis, each claimed to foreshadow a continuation or reversal.
This is the case where balance matters most, because the evidence is genuinely mixed, not empty. Lo, Mamaysky & Wang (2000), in one of the more rigorous studies of the field, used a systematic kernel-smoothing approach to detect patterns automatically and found that several of them carry marginal statistical information about subsequent returns. That is a real, if modest, positive result.
The caveats are what move this into the graveyard. The information found was small, its tradeable value after costs was limited, and the study removed the very thing most retail traders rely on — human, subjective identification. In practice two traders draw the same neckline or handle differently, so live results are not reproducible the way the automated test was.
So the honest verdict is narrow: some patterns are not noise, but there is no peer-reviewed, out-of-sample evidence of a robust edge after costs, and broad reviews like Park & Irwin (2007) reach the same cautious place. The burden of proof for a deployable edge has not been met — even though, unlike astrology, the kernel of information here is real.
Sources
- Lo, Mamaysky & Wang (2000), "Foundations of Technical Analysis", Journal of Finance
- Park & Irwin (2007), "What Do We Know About the Profitability of Technical Analysis?", Journal of Economic Surveys
Frequently asked
Do classic chart patterns work in 2026?
The evidence is mixed and weak rather than zero. Lo, Mamaysky & Wang (2000) found that some patterns carry marginal statistical information, but the tradeable value after costs is limited, and pattern identification is subjective across analysts. There is no peer-reviewed, out-of-sample evidence of a robust edge after costs — but it would be wrong to say these formations contain no information at all.
Is head-and-shoulders or cup-and-handle trading profitable?
Not reliably after costs, on the published evidence. The marginal information some studies detect tends to shrink or disappear once spreads, slippage and subjective identification are factored in. Where two traders draw the neckline or handle differently, results stop being reproducible. The patterns are a real object of study; they are not a validated, cost-surviving edge.
Not investment advice — your mileage may vary, but the burden of proof is on the person claiming an edge. This entry describes general research and published evidence (or its absence), not a recommendation. See the full disclaimer.