mercatusdata

Does a stock near its 52-week high or low keep going? 35 large US stocks, 1962–2026

Does reaching the top of its 52-week range say anything about how a stock does next? What about the bottom? Across 35 large US stocks since 1962, the honest answer is: not much. After moving into the top 5% of its range, a stock went on to trail its own usual pace by 0.18 percentage points on average over the next 20 sessions. After moving into the bottom 5%, it beat that pace by 0.58 points. Next to how much individual outcomes swing, both numbers are small.

−0.18

points of excess over 20 sessions after entering the top 5% of the 52-week range · 48% of 3,282 episodes beat drift · statistically distinguishable from zero

+0.58

points of excess over 20 sessions after entering the bottom 5% of the 52-week range · 52% of 1,150 episodes beat drift · not clear of noise

The short answer

Near the high, 48% of 3,282 episodes beat their stock’s own drift over the next 20 sessions. Near the low it was 52% of 1,150. Either way, that is close to a coin toss. The small lag after a high does hold up statistically. The gain after a low does not, and since 1990 it has shrunk to +0.09 points across 800 episodes.

What was measured

The signal is where a day’s close sits within the stock’s range over the last 252 sessions, about a year, from the lowest daily low to the highest daily high. A close at 95% of that range or higher counts as near the 52-week high; one at 5% or lower counts as near the low. A stock can be near its high without setting a new one.

An episode starts on the first day a stock enters one of those zones. Entries within 10 sessions of each other count once, so a stock that hovers near its high for weeks is not counted again and again.

Each episode is followed for 1, 5 and 20 sessions. The return over each horizon is compared with the stock’s own average return over the same horizon, worked out from the stock’s whole price history up to and including the signal day, once at least 30 such returns exist. That average is what this note calls the stock’s drift, or its usual pace. Subtracting it leaves the part of the move that the stock’s normal tendency to rise or fall does not explain. It is not a comparison with the market.

The sample is 35 large US stocks, chosen in 2026 as the large US companies people most often look up, across eight sectors, before any of these results were computed; each is followed from its first available price, 1962 to 2026: 3,282 episodes near the high and 1,150 near the low with a full 20 sessions of data after them.

Results

Excess return over own drift, percentage points
Position in 52-week rangeEpisodes (20 sessions)1 session5 sessions20 sessionsMedian, 20 sessionsSpread (sd), 20 sessionsBeat own drift, 20 sessionst-statistic (monthly averages)
Top 5%3,2820.00−0.09−0.18−0.319.548%−3.6
Bottom 5%1,150+0.22+0.51+0.58+0.3512.752%+1.7

One thing to keep in mind when reading the t-statistics: episodes bunch together. In a broad rally, many stocks sit near their highs in the same weeks; in a sell-off, many sit near their lows. Counting each episode as independent would overstate the evidence, so the test runs on monthly averages instead. On that basis the lag near the high (t −3.6) is statistically distinguishable from zero. The gain near the low (t +1.7) is not.

Either way, the averages are tiny compared with the spread. The standard deviation of 20-session excess returns is 9.5 points near the high and 12.7 near the low, while neither average reaches one point.

No consistent direction by decade

Split by decade, the picture keeps changing. For all companies in the sample, the average near the high moves between −0.78 and +0.66 points, and the average near the low between −1.13 and +3.72. The largest figure, in the 1980s, rests on just 138 episodes.

Same 10 companies, present since the 1960sAll companies in the sample at the time1960s, same 10 companies: −0.18 points over 157 episodes−0.181960s, all companies in the sample at the time: −0.18 points over 157 episodes−0.181960s1970s, same 10 companies: +0.07 points over 181 episodes+0.071970s, all companies in the sample at the time: −0.53 points over 210 episodes−0.531970s1980s, same 10 companies: +0.21 points over 245 episodes+0.211980s, all companies in the sample at the time: −0.40 points over 469 episodes−0.401980s1990s, same 10 companies: +0.04 points over 265 episodes+0.041990s, all companies in the sample at the time: +0.66 points over 588 episodes+0.661990s2000s, same 10 companies: −0.38 points over 188 episodes−0.382000s, all companies in the sample at the time: −0.78 points over 498 episodes−0.782000s2010s, same 10 companies: −0.95 points over 235 episodes−0.952010s, all companies in the sample at the time: −0.65 points over 842 episodes−0.652010s2020s, same 10 companies: +0.55 points over 116 episodes+0.552020s, all companies in the sample at the time: +0.56 points over 518 episodes+0.562020sexcess return, percentage points
Near the 52-week high: average excess return 20 sessions after entering the top 5% of the range, by decade.
Same 10 companies, present since the 1960sAll companies in the sample at the time1960s, same 10 companies: −0.29 points over 67 episodes−0.291960s, all companies in the sample at the time: −0.29 points over 67 episodes−0.291960s1970s, same 10 companies: −0.42 points over 111 episodes−0.421970s, all companies in the sample at the time: +0.69 points over 145 episodes+0.691970s1980s, same 10 companies: +1.48 points over 59 episodes+1.481980s, all companies in the sample at the time: +3.72 points over 138 episodes+3.721980s1990s, same 10 companies: +0.97 points over 55 episodes+0.971990s, all companies in the sample at the time: +0.29 points over 153 episodes+0.291990s2000s, same 10 companies: −1.74 points over 86 episodes−1.742000s, all companies in the sample at the time: −1.13 points over 277 episodes−1.132000s2010s, same 10 companies: +1.22 points over 57 episodes+1.222010s, all companies in the sample at the time: +1.69 points over 166 episodes+1.692010s2020s, same 10 companies: +1.89 points over 67 episodes+1.892020s, all companies in the sample at the time: +0.32 points over 204 episodes+0.322020sexcess return, percentage points
Near the 52-week low: the same measure after entering the bottom 5% of the range.

The solid bars follow only the 10 companies that have been in the sample since the 1960s (Boeing, Caterpillar, Chevron, Disney, Johnson & Johnson, Coca-Cola, Lockheed Martin, McDonald’s, Procter & Gamble, Exxon Mobil), so a fixed group can be set against a sample that grew from 10 companies to 35.

Stock by stock

Individual stocks differ widely. Among the 33 with at least 30 episodes near the high, averages run from +4.20 points for Advanced Micro Devices to −3.06 for Mastercard, and 15 of the 33 have a positive average. With that many stocks, some will look extreme by chance, so the best and worst figures probably overstate how different those companies really are.

all episodes: −0.18Advanced Micro Devices (68)Advanced Micro Devices: +4.20 points over 68 episodes, 53% beat drift+4.20Amazon (62)Amazon: +2.01 points over 62 episodes, 44% beat drift+2.01Apple (99)Apple: +1.66 points over 99 episodes, 57% beat drift+1.66Bank of America (93)Bank of America: +0.99 points over 93 episodes, 54% beat drift+0.99Lockheed Martin (129)Lockheed Martin: +0.98 points over 129 episodes, 53% beat drift+0.98Meta Platforms (44)Meta Platforms: −1.48 points over 44 episodes, 39% beat drift−1.48UnitedHealth (128)UnitedHealth: −1.63 points over 128 episodes, 44% beat drift−1.63Home Depot (94)Home Depot: −1.84 points over 94 episodes, 43% beat drift−1.84Walmart (118)Walmart: −2.10 points over 118 episodes, 36% beat drift−2.10Mastercard (52)Mastercard: −3.06 points over 52 episodes, 25% beat drift−3.06… 23 other stocks lie betweenexcess return, percentage points
Near the 52-week high: average excess return 20 sessions after entering the top 5% of the range, the five best and five worst stocks; episodes in brackets; solid bars mark the highest and lowest.

Near the low, only 19 stocks reach 30 episodes. Their averages run from +3.61 for JPMorgan Chase to −2.20 for Boeing, and 13 are positive.

What this does not say

This is a record of what followed in the past, not a forecast or a trading rule, and it says nothing about the next episode.

The sample is made of survivors. All 35 companies are large today, so a company that sank near its 52-week low and never recovered is not here. Leaving such companies out flatters the results near the low.

The comparison is with each stock’s own drift, not with the market: a positive number means a stock did better than its own usual pace, not that it beat an index. And no costs are deducted, whether commissions, bid–ask spreads or taxes.

Data and method