Factor Investing

The 52-Week High Effect: Why Stocks Near Their Highs Often Keep Winning

The 52-week high effect is one of the cleanest ideas in price-based investing: stocks trading close to their 52-week highs tend to outperform stocks far below them. Not “cheap stocks bounce.” Not “buy the dip.” The signal says the opposite: strength near prior highs contains information.

14 min readAXLFI Blog

Overview

The 52-week high effect is one of the cleanest ideas in price-based investing: stocks trading close to their 52-week highs tend to outperform stocks far below them.

That is the core result. Not “cheap stocks bounce.” Not “buy the dip.” The signal says the opposite: strength near prior highs contains information.

The key academic paper is George and Hwang (2004), who showed that a stock’s nearness to its 52-week high helps explain momentum profits and predicts future returns. Their central claim is not just that recent winners keep winning, but that proximity to the 52-week high itself captures a large part of the effect.

52-week high proximity

pth(t) = P(t) / H(t)²⁵²

Core prediction

high pth ⇒ higher expected future returns

Visual

Price and Rolling 52-Week High

The signal in one picture. The stock price is plotted alongside its rolling 52-week high. When the price is close to the high (shaded zone), the stock scores high on the proximity measure. That zone is where the 52-week high effect predicts outperformance.

pth(t) = P(t) / H²⁵²(t)

Visual

Bucketed Forward Returns by Proximity

The strongest evidence-style chart for the effect. Stocks are split into buckets by how close they are to their 52-week high, then average forward 6-month returns are plotted. The pattern is clear: stocks nearest their highs delivered the strongest subsequent returns.

E[R(t+6m) | pth bucket] increases as pth → 1

Visual

The Psychological Barrier

The behavioral explanation in one diagram. A stock drops from a prior high, recovers toward that level, and investors hesitate — anchoring on the old high as a ceiling. When the stock finally pushes through, the underreaction reverses and the stock drifts higher. That hesitation is the friction the 52-week high effect exploits.

Article Section

What the signal actually is

A simple way to define the signal: pth(t) = P(t) / H²⁵²(t), where P(t) is today’s price and H²⁵²(t) is the highest price in the last 52 weeks. If pth(t) ≈ 1, the stock is trading near its 52-week high.

Stocks with higher pth(t) values tend to do better going forward than stocks with low pth(t) values. George and Hwang’s paper made this variable famous by showing it predicts cross-sectional stock returns and helps account for conventional momentum profits.

Signal

pth(t) = P(t) / H²⁵²(t)

Near high

pth(t) ≈ 1 ⇒ within striking distance bullish

Far from high

pth(t) ≪ 1 ⇒ lagging badly bearish

Article Section

Why this is interesting

At first glance, the effect sounds backwards. Most people instinctively think: near high ⇒ overbought. The 52-week high effect says that, on average, that intuition is often wrong. A stock near a major prior high may actually be underreacted to, not overdone.

The usual behavioral explanation is that investors anchor on the old high as a salient reference point. If the stock climbs back toward that level, investors may still react sluggishly, treating the old high as a psychological barrier instead of fully updating to the new information in the price.

near the high ≠ overbought — it often means underreacted

Article Section

The economic intuition

A clean way to think about it: new information arrives, the price rises toward the prior high, investors underreact because they anchor on the old high as a ceiling, and the stock continues to drift higher as the market slowly adjusts.

The old high acts as a mental anchor. Instead of immediately repricing the stock based on new fundamentals, some investors seem to behave as though the prior high itself is meaningful. That slows full price adjustment.

That is why this effect is more interesting than a generic breakout rule. It is trying to exploit a specific behavioral friction: anchoring to salient price levels.

new informationprice rises toward prior highinvestors underreactcontinued drift

Article Section

A practical way to use it

A simple workflow: screen the universe, rank by nearness to 52-week high, filter for quality, and hold leaders. A basic score is score(i) = P(i) / H²⁵²(i). Then focus on the top bucket.

A score ≥ 0.90 means the stock is within 10% of its 52-week high. A more selective version might focus only on names at ≥ 0.95. Then combine with other filters like liquidity, trend above a long moving average, earnings strength, and sector leadership.

screenrank by nearness to highfilterhold leaders

Basic score

score(i) = P(i) / H²⁵²(i)

Moderate filter

score ≥ 0.90 within 10%

Selective filter

score ≥ 0.95 within 5%

Article Section

Why the effect is not the same as a breakout

This is an important distinction. A breakout rule usually says: buy if P(t) > H²⁵²(t). The 52-week high effect is broader. It says even approaching the high can matter.

So the anomaly is not limited to the exact breakout day. It is about distance to the high as an informative variable. That makes it useful for ranking and stock selection, not just signal timing.

the signal is distance to the high, not the breakout itself

Breakout rule

buy if P(t) > H²⁵²(t) binary

52-week high effect

P(t) ↑ H²⁵²(t) ⇒ higher expected return continuous

Article Section

Why it may work across markets

Research beyond the original U.S. study has found similar profitability in many international equity markets. One international study reports that the 52-week high momentum strategy produced profits in 18 of 20 markets and statistically significant profits in 10 of them, while also existing independently of standard momentum measures.

That matters because it suggests the effect is not just a quirky U.S. artifact. If investor anchoring to prior highs is a universal behavioral trait, you would expect the signal to appear anywhere equity markets exist.

International evidence

profitable in 18 of 20 markets robust

Significant

statistically significant in 10 of 20

Article Section

The downside of the strategy

Like most momentum-style effects, the 52-week high effect is not smooth. It tends to struggle when leadership reverses abruptly, markets experience violent mean reversion, crowded winners unwind, or sharp macro turning points hit momentum trades.

The payoff structure is similar to other momentum strategies: a moderate win rate combined with average winners that exceed average losers. The effect can work well over time and still suffer painful reversal periods.

momentum reversals are the price of admission — risk controls are essential

Payoff structure

E[R] = p · W̄ − (1 − p) · L̄

Key property

p not necessarily high, but W̄ > L̄ positive skew

Article Section

The deeper insight

The most interesting part of the 52-week high effect is that it says price levels matter, not just returns. That is subtle.

Standard momentum mostly looks at past performance: P(t)/P(t−k) − 1. The 52-week high effect looks at where price sits relative to a salient historical anchor: P(t)/H²⁵²(t). That means it is not only about trend persistence. It is also about investor psychology around prior peaks.

This is what makes the anomaly conceptually rich. It is trying to capture something about how investors think about price levels, not just price changes.

Standard momentum

P(t) / P(t−k) − 1 returns-based

52-week high effect

P(t) / H²⁵²(t) level-based

Article Section

A practical rule set

A straightforward implementation following six clean steps. This is not the only way to do it, but it is a clean way to turn the idea into a usable selection model.

Compute nearness to 52-week high

pth(i) = P(i) / H²⁵²(i)

For each stock, calculate pth(i) = P(i) / H²⁵²(i). This is the core signal.

Keep only liquid names

avg dollar volume > threshold

Filter for stocks with sufficient average dollar volume. This avoids illiquid names that are hard to trade and subject to more noise.

Rank stocks cross-sectionally

pth_rank(i) = percentile rank of pth(i)

Compute the percentile rank of each stock’s pth value relative to the universe. This converts the raw ratio into a comparable score.

Buy the top bucket

pth_rank(i) ≥ 90th percentile

Focus on stocks at or above the 90th percentile of pth rank. These are the names closest to their 52-week highs.

Add an absolute trend filter

P(t) > MA(200)

Require price to be above the 200-day moving average. This removes stocks with high proximity but negative absolute trends.

Rebalance monthly

Monthly rebalancing is often more stable than daily reactions. Replace stocks that drop out of the top bucket with new leaders.

Article Section

The 52-week high philosophy

The best summary of the 52-week high effect captures why this anomaly has remained a focus of both academic research and practical investing.

strength near prior highs contains information
investors anchor to salient price levels
underreaction near highs creates persistent drift
proximity to the high explains a significant part of momentum

Conclusion

Why the framework still holds up

The 52-week high effect is one of the strongest examples of a market rule that feels counterintuitive but has deep logic behind it. Stocks closest to their prior highs often remain the market’s leaders.

The reason is not that highs are magical. It is that investors seem to anchor to salient price levels and underreact when strong stocks press back toward those highs. That specific behavioral friction is what makes nearness to the 52-week high a persistent and well-documented signal.

That is why nearness to the 52-week high has remained an important variable in both the academic momentum literature and practical growth investing. The effect says something fundamental about how investors process price levels — and that insight alone makes it worth understanding.