Among the least traded shares, do the week's biggest risers trail its biggest fallers the next week?
Study 58 · Changed a rule or plan · recorded 18 October 2026 · answer: Yes · full paper with figures
Correction, 19 October 2026. This reading was decided by the clipped fallback measure, whose false-alarm rate was not checked on this study's own design. The plain measure gives the same answer, so the reading stands; the note records that the deciding measure was untested here.
The question
In a documented weekly reversal, a week's biggest risers do worse the next week than its biggest fallers. Our scans of 2013-2017 saw it only among the least traded third of shares; we tested it on 2018-2022, years earlier studies had used for other questions.
Rules fixed before the result
Fixed in advance: at least 200 weeks of at least 200 shares each; the reversal holds when the range from the next week's first opening price lies wholly below zero, read against -0.63 points in 2013-2017, about half the -1.11 from the last close. The plain average decided unless its 95% range was wider than 0.88 points; it was 0.89 wide, so the clipped average, the measure we fall back on, decided, and no minimum or bar was changed.
Data and method
Each week, among shares priced at 5.00 or more in US currency in the least traded third by the value of shares traded, we compared the week's top fifth by return with its bottom fifth over the next week, from its first opening price, the week's last close or its first close, over 249 weeks, beside the other thirds. Fixed in advance: two other builds of that third must agree in direction, and untraded shares must not carry the result.
The answer
Yes. In 2018-2022, among the least traded third, the top fifth's next-week return less the bottom fifth's, from the next week's first opening price, was -0.51 percentage points (95% range -0.86 to -0.16), wholly below zero as required, against -0.63 in 2013-2017. From the week's last close it was -0.88 (-1.24 to -0.53) and from the next week's first close -0.62 (-0.92 to -0.33). Among the middle third it was -0.15 (-0.53 to 0.23) and among the most heavily traded third -0.37 (-0.76 to 0.03), neither wholly below zero. The two other ways of building that third agreed in direction (-0.38 and -0.47), shares traded on both days that matter gave -0.51, and the reversal after heavy-trading weeks less that after light ones was -0.02 (range -0.52 to 0.47, allowing a negative difference as large as the main finding).
Main estimate
-0.005 fraction (95% range -0.009 to -0.002), from 249 observations
What it changes
By the rule set in advance, our studies of returns around events on thinly traded shares will add a control for this weekly reversal, and the liquidity questions have their first finding within our development data. The next liquidity question, whether moves in the most traded shares lead those in the least traded, is its natural successor when its turn comes. The plan expected this answer with medium confidence, at -0.3 to -0.7 points.
What remains uncertain
These years' prices were read before by studies eighteen, thirty-two and fifty-seven for other questions, so this is a check within our development data, not a fresh test. The data carry no quoted bid or ask prices, the weeks are read as independent with equal weights, and the price filter uses the adjusted close. The difference was largest in 2020 (-1.93 points, range -2.93 to -0.94) and near zero in 2018, 2019 and 2022.
How it was made
Run took 6 minutes; independent code review: approve; the recorded calculation reruns to the same result.
Fingerprints
plan ff60dfa6019fc5b5040f0dcc1f9ed6b2e6947380ab610b3c75806a2d992bafcb result c25c771ca6c6a6f0adad44a3eb54dabb33a9adad98c7141368fe6beb9aa466d3
Exploratory work on development data; not a performance claim.