Do low-priced shares fare differently from comparable companies in the year after a large one-day fall?
Study 48 · Changed a rule or plan · recorded 11 October 2026 · answer: Partly · full paper with figures
The question
Low-priced here means a last share price before the fall of at least 1 and under 10 in US currency, and a large one-day fall is a drop of 10% or more. We compare each fallen share with five comparable companies of the same industry and similar trading size, chosen by their trading before the fall.
Rules fixed before the result
Fixed in advance: at least 1,000 falls by 500 companies, and a comparison able to detect 0.10, about 9.5%, in the typical holder's wealth; it could detect 0.053. A lag lasts if the year's whole range lies below zero and reaches a 0.10 shortfall; it is specific to falls if ordinary dates show none, and survives the volatility adjustment if approximate average wealth is behind. It had to withstand a further 30% loss on shares leaving the market after halving.
Data and method
We rebuilt study eighteen's falls and kept at most one per company in any year, leaving 1,916 falls by 739 companies. On the logarithmic scale of the typical holder's wealth the rare very large rises no longer dominate the average and a more volatile share grows more slowly; approximate average wealth adds that back, so the two measures can differ. The same companies' ordinary dates are quarter starts with no one-day fall of 10% or more in the year before.
The answer
Partly. Over the year after the fall, the typical holder's wealth in fallen shares ended about 17.5% below that in comparable companies (95% range 13.4% to 21.6%), a shortfall beyond the 9.5% named in advance, most of it after the first quarter; on the same companies' ordinary dates it showed no measurable gap. These shares were typically 1.36 times as volatile as their comparables, and approximate average wealth was not measurably behind, from 0.7% less to 10.2% more, so here the two measures differ. Of twelve checks fixed in advance, ten left the typical holder's shortfall in place; the two whose ranges include zero are approximate average wealth and a difference of returns with extreme years pulled in. Valuing shares that left the market after losing half their value at a further 30% loss did not change the answer.
Main estimate
-0.175 fraction (95% range -0.216 to -0.134), from 1916 observations
What it changes
By the rule fixed in advance, the shortfall in the typical holder's wealth did not persist under the volatility adjustment, as approximate average wealth was not measurably behind, so no year-long expectation for such falls is proposed and the question of the year after them closes with that reason. The plan had expected, with low confidence, that the shortfall would persist under the adjustment; it did not.
What remains uncertain
Studies eighteen, thirty-two and thirty-five read these falls' first days before this plan was fixed; same-date falls share market shocks the ranges treat as independent, though resampling by calendar month agreed. The volatility adjustment is approximate: it assumes roughly lognormal returns, which fat tails break, and uses each fall's own noisy variance over the year. In the typical holder's wealth the last period (2020 to 2021) alone was not measurably behind; comparables were not matched on price level or distress.
How it was made
Run took 6 minutes; independent code review: approve; the recorded calculation reruns to the same result.
Fingerprints
plan 0548d23c8193f51f961bc652ab86c0491dc5f7f1f463135aa7a587f72bdb7b43 result 4677537e0e66bea1cfe9d23f1938158f3b20fd24af858786406205da1d82e1fc
Exploratory work on development data; not a performance claim.