Correction, 30 September 2026. Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged.
Correction, 1 October 2026. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan.
Correction, 19 October 2026. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
The question
We ask how many future splits that increase the number of shares, not reverse splits, a test would need to detect a planted fall of 1.75 percentage points in five-session returns at least 80% of the time. The answer sets the smallest sample any later plan to test for such a fall may use.
Rules fixed before the result
The rules for each possible answer were written down and fixed before any result was seen.
Data and method
Each split session was replaced by an ordinary session of the same company, and a fall of 1.75 percentage points was planted in the five-session returns. In 25 random groups of 64, 47, 32, 24 and 16 companies, and in all 94, we counted over thousands of repetitions how often a rank-based test and a test on average differences with simulated thresholds detected the fall. A second version widened the spread of differences to match that seen in real splits.
What remains uncertain
Whether future events will behave like past ones is still open: any count of years assumes splits keep coming at the rate seen in our development years. The planted fall of 1.75 percentage points is the estimate from those years itself, so planning on it is optimistic. The simulation draws ordinary sessions independently, so it does not treat companies that move together on the same dates.