Do the least traded shares, unlike the most traded, reverse by at least a tenth of a percentage point overnight?
Study 93 · Changed a rule or plan · recorded 9 November 2026 · answer: Partly · full paper with figures
The question
Among shares priced five or more at the previous session's close, the least traded third by recent trading value, with a usable next day of prices, we measure how much more the day's biggest fallers than its biggest risers gained the next day. Is it at least 0.10 percentage points, overnight, and absent in the most traded third?
Rules fixed before the result
Fixed before any outcome, on the 95% range of the least traded third's reversal: wholly above zero and reaching 0.10 percentage points gives yes if also overnight, accounted for by where the close sat in the day's range, absent in the most traded third and below the estimated gap everywhere, else partly; above zero but under 0.10, partly. Under 0.10 and including zero gives not at the size tested; anything else, or a failed floor or power check, not settled.
Data and method
Each session, shares priced five or more at the previous session's close, with a usable next day of prices, are split into thirds by recent trading value. In the least traded third, the next-day return against the market fund of the fifth that fell most, less the fifth that rose most, is averaged, its 95% range allowing for correlated nearby sessions. The buyer-seller price gap is estimated from daily highs and lows; earlier years were resampled to check false alarms.
The answer
Partly. Over 1,258 sessions from 2018 to 2022, the day's biggest fallers in the least traded third gained 0.209 percentage points more than its biggest risers the next day (95% range 0.137 to 0.281), wholly above the 0.10 set in advance. The reversal was carried overnight and by where the close sat in the day's range; the part in the next day's trading hours could not be told from zero. It stayed below the gap between buyers' and sellers' prices, as estimated from daily highs and lows, in each of nine groups by trading value, price and that estimate. The most traded third's 95% range, from fallers gaining 0.107 points less than risers to 0.058 more, rules out a reversal of 0.10 but not a move that size the other way, so its absence was not shown.
Main estimate
0.209 percentage points (95% range 0.137 to 0.281), from 1258 observations
What it changes
Under the rule fixed in advance, the least traded shares' next-day reversal is recorded as overnight, taken by where the close sat in the day's range and below the gap as estimated from highs and lows, with the most traded third left open. Later reversal studies set out in advance how the reversal compares with that estimated gap and which shares count. No trading decision changes.
What remains uncertain
These years were used by earlier studies, so this is a check inside our development data, not an independent test, and sessions varied more than the plan allowed for. Side readings were each read alone: with no effect beyond the earlier years' averages, at least one shows a pattern in 29.8 percent of resampled runs. The gap is estimated, not quoted, so staying below it fits a bounce between buyers' and sellers' prices without showing one.
How it was made
Run took 64 minutes; independent code review: approve; the recorded calculation reruns to the same result.
Fingerprints
plan b597012a7fc6484e7074ef663371af3143d10385398651fb20e7c7534d60f047 result b43fc07fb178ce5e0afd2f6e660a239e8a1519da9d26e85d63aef4479c6185d5
Exploratory work on development data; not a performance claim.