52-Week-High Momentum
Buy stocks trading near their 52-week high, betting the anchor effect carries them higher.
Documented strong in-sample by George & Hwang (2004), but exposed to severe momentum crashes and the post-publication decay common to anomalies.
- Why it fails
- The signal rides momentum, which periodically suffers violent reversals (momentum crashes), and the in-sample edge has thinned in the way published predictors typically do after they are publicized.
- When / how it stopped
- George & Hwang (2004) documented the effect strongly in-sample, but it shares momentum's vulnerability to crashes (Daniel & Moskowitz, 2016) and the broad post-publication decay catalogued by McLean & Pontiff (2016).
The 52-week-high strategy is an elegant, behaviorally motivated form of momentum: buy the stocks trading closest to their trailing 52-week high, on the theory that the high acts as a psychological anchor. Investors hesitate to bid a stock past a salient recent peak, news that should push it higher gets underreacted to, and the price drifts up as the anchor gives way.
George & Hwang (2004) documented this effect strongly in-sample in the Journal of Finance, and notably found that proximity to the 52-week high often predicted future returns better than past returns themselves — a clean, intuitive result that made it a favorite screen.
The problem is what the strategy inherits. It is fundamentally a momentum trade, and Daniel & Moskowitz (2016), in “Momentum Crashes”, document that momentum portfolios are exposed to rare but severe reversals — concentrated in panic-driven rebounds where previously-beaten stocks surge and momentum positions take heavy losses. A 52-week-high screen, by construction, sits on the wrong side of exactly those moves.
Layered on top is the general decay McLean & Pontiff (2016) measured across published anomalies: once a predictor is in the literature, arbitrage tends to erode it.
None of this means the anchor effect was never real. It means the smooth backtest hides a fat tail and a fading premium. The reader can weigh whether that trade-off is acceptable.
Sources
- George & Hwang (2004), "The 52-Week High and Momentum Investing", Journal of Finance
- Daniel & Moskowitz (2016), "Momentum Crashes", Journal of Financial Economics
- McLean & Pontiff (2016), "Does Academic Research Destroy Stock Return Predictability?", Journal of Finance
Frequently asked
Does 52-week-high momentum still work in 2026?
Its track record is mixed. George & Hwang (2004) found that nearness to the 52-week high was a strong predictor of future returns in-sample — often a better one than past raw returns. But the strategy is a flavor of momentum, and Daniel & Moskowitz (2016) document that momentum suffers infrequent but devastating "crashes," typically in market rebounds after sharp declines. Combined with the general post-publication decay catalogued by McLean & Pontiff (2016), the realized edge is far less reliable than the headline in-sample numbers imply.
Why is the 52-week-high effect risky despite strong backtests?
The effect is built on the same engine as price momentum, and that engine breaks down badly at certain points in the cycle. Daniel & Moskowitz (2016) show momentum portfolios can lose a large fraction of their value in short bursts when beaten-down stocks violently rebound — exactly when a 52-week-high screen is least exposed to them. So a smooth-looking average return masks a tail that can erase years of gains.
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.