The research record, in plain text

Every study the research system has chosen, designed, run and published, newest first, including the ones that stopped or settled nothing. Read from the record on 13 November 2026. The interactive version is the Research page.

96 studies published. 34 of the 96 ended inconclusive or stopped by their own checks, and are published anyway. No result here has been confirmed on data set aside for confirmation, and none has been reviewed by an independent expert.

How the studies ended

Outcome	Studies
Concluded	4
Not supported	2
Changed a rule or plan	53
Checked a method	3
Inconclusive	19
No reading (stopped by its own checks)	15

Results later corrected

A corrected study keeps its original registration and result on record; the correction is dated beside it.

Study 58: Note added 18 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.
Study 57: Note added 18 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 would have read the pattern as a closing-price artefact, so the reading rests on the clipped measure, which was untested here. Corrected 7 November 2026. The answer said the most heavily traded third's overnight difference was above zero, but only by the clipped measure, whose range cleared zero by 0.03 points. From 31 October 2026 that measure no longer decides two-group comparisons in new designs, so the sentence gives the plain average, whose range holds zero. The overall Yes remains the clipped measure's reading; as noted on 18 October, the plain measure would read the pattern as a closing-price artefact.
Study 55: Note added 18 October 2026. This reading was decided by the fallback measure, which pulls in the most extreme 1% of values at each end. Its false-alarm rate was checked on this study's own design and stayed within the limit the registration allows, and the plain measure gives the same answer. The reading stands.
Study 50: Note added 20 October 2026. This study's answer was changed from Yes to Not settled. Its shortfall rested on a clipped measure that a later check (study 56) found reports a difference that is not there too often on this design. A re-test of the same notices (study 60), on a measure whose false-alarm rate holds, did not establish the shortfall and did not rule out a shortfall of about three points. The registration and the original result are unchanged.
Study 34: The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 33: The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 32: The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 31: The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 30: The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 29: The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 28: The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged. The outcome label was changed from "concluded" to "not supported", which states more exactly what the study found: the idea it tested, switching to the market-wide rule, was not supported.
Study 27: The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. The text of this record was rewritten to the project's editorial standard, from the study's plan, result and records rather than its earlier wording. What the study found is unchanged.
Study 26: The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was written and reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 25: This text was written and reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 24: This text was written and reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 23: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 22: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 21: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 20: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 19: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 18: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 17: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 16: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 15: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 14: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 13: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 12: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 11: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 10: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 9: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 8: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 7: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 6: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 5: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 4: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 3: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 2: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. The outcome label now follows the study's value record, which counts this study as having changed a rule or a plan. This text was written and reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.
Study 1: Technical wording rewritten in shorter sentences under the readability rule; the meaning is unchanged. This text was written and reviewed after the study finished, in a catch-up of the earliest studies; the result, outcome and dates are unchanged.

All studies

Study 96 · Changed a rule or plan · recorded 12 November 2026
Does trading volume rise more with recent gains in shares trading above their past year's average price?
Study 95 · Changed a rule or plan · recorded 12 November 2026
Do shares' daily lows pile up just past ten percent below the previous close, where short selling is limited?
Yes: from 2018 to 2022, shares' daily lows piled up just past ten percent below the previous close, by at least the size set in advance, and not at nearby levels. Later designs may treat that point as a boundary.
Study 94 · Concluded · recorded 11 November 2026
Across their own recent gains, does trading volume rise more steeply in the least traded shares than in the most traded?
Partly: across their own recent gains, trading volume rose more steeply in the least traded third of shares than in the most traded, by 0.69 (95% range 0.51 to 0.87) against a bar of 0.31; the loss side stayed unresolved.
Study 93 · Changed a rule or plan · recorded 9 November 2026
Do the least traded shares, unlike the most traded, reverse by at least a tenth of a percentage point overnight?
Partly. Among the least traded third of shares with a usable next open and close, the day's biggest fallers gained 0.209 percentage points more than its biggest risers the next day (95% range 0.137 to 0.281), carried overnight.
Study 92 · Not supported · recorded 9 November 2026
Third run: among a day's biggest fallers, do shares closing at a year's low rank lower the next day?
Not at the size tested. Among the day's biggest fallers, those at a year's low ranked 0.003 lower next day on a 0 to 1 scale (95% range 0.004 higher to 0.009 lower), under 0.012, the smallest gap worth reporting.
Study 91 · No reading (stopped by its own checks) · recorded 8 November 2026
Run again: among a day's biggest fallers, do shares closing at a year's low rank lower the next day?
The rerun stopped before measuring any return in 2018 to 2022: one figure of the earlier attempt's checks did not match, because it depends on the smallest gap worth reporting, which the rerun raised from ten to twelve thousandths.
Study 90 · No reading (stopped by its own checks) · recorded 7 November 2026
Among a day's biggest fallers, do shares that closed at their lowest price of the past year rank lower the next day?
The study stopped before measuring any return in 2018 to 2022: too few sessions in those years had five or more large fallers closing at a year's low for the planned test to have the power it needed.
Study 89 · Inconclusive · recorded 6 November 2026
How often do our two-group decision rules report a false gap when one group's extreme values reach further?
The corrected study settled neither family of rules: study 67's widened rule reported false gaps on one side in 3.3% of comparisons, too near the 3.75% limit to call, so its earlier uses stand with that named.
Study 88 · No reading (stopped by its own checks) · recorded 5 November 2026
How often do our two-group decision rules report a false gap when one group's extreme values reach further?
Study 86's corrected re-run stopped before reading any data, so no rate was measured: a check in the plan demands more than the plan's method can deliver. The owner has reopened the question for a corrected plan.
Study 87 · Changed a rule or plan · recorded 5 November 2026
Do the day's biggest risers closing at their highest price of the past year rank higher on the next day's return, chiefly overnight?
Partly. Among the day's biggest risers, those closing at their highest price of the past year ranked 0.018 higher next day (0.010 is one percentile point; 95% range 0.012 to 0.023); which part of the day carries it is unresolved.
Study 86 · No reading (stopped by its own checks) · recorded 4 November 2026
How often do our two-group decision rules report a false gap when the groups' extreme values differ?
The study stopped at a check planned before any rate was measured, so a corrected design is to follow: rebuilt as planned, one earlier study's decision rule did not return the count of false gaps that study's record holds.
Study 85 · Changed a rule or plan · recorded 4 November 2026
Does a share that beat similar shares rank lower in the next day's trading hours, apart from where it closed?
Partly: in this sample, shares that beat similar shares most ranked 0.0045 lower on next-day trading-hours returns than those trailing most (scale 0 to 1; 95% range 0.0004 to 0.0086); whether this holds apart from where they closed is unresolved.
Study 84 · Changed a rule or plan · recorded 4 November 2026
Is the next-day reversal weaker in shares whose daily returns tend to continue, apart from where the close sat?
Partly; a note is proposed. In shares trading at the next open and close, a 0.226 percentage-point smaller reversal where returns tended to continue than reverse (95% range 0.169 to 0.283) went with the close's place in the day's range.
Study 83 · Changed a rule or plan · recorded 3 November 2026
Do shares that close near the day's high earn less the next session than shares that close near the low?
Partly: the next session's return was 0.247 percentage points lower after a close at the day's high than after one at its low (95% range 0.176 to 0.318), beyond the 0.10 set in advance; its source stayed open.
Study 82 · No reading (stopped by its own checks) · recorded 3 November 2026
Do shares that close near the day's high earn less the next session than shares that close near the low?
The study stopped before measuring any return in 2018 to 2022: its false-alarm check called for a wider interval than the plan allowed, which would have left too little power to detect a gap of 0.09 percentage points.
Study 81 · Changed a rule or plan · recorded 2 November 2026
Do shares that close just below a whole-number price earn less the next session than shares just above?
Yes: from 2018 to 2022, shares closing one to five hundredths below a whole-number price earned 0.12 percentage points less the next session than those just above (95% range 0.086 to 0.156); around prices ending in .50 it was smaller.
Study 80 · Inconclusive · recorded 1 November 2026
Do recording errors in the payout amounts in our price data carry study 77's gap between payout-yield thirds?
Not settled: on the payout days we could compare, under half of each third's sample, most recorded payouts matched the declared amount, but the range for the errors' share of study 77's gap still reaches half of it.
Study 79 · No reading (stopped by its own checks) · recorded 1 November 2026
Does the payout amount in our price data match the amount each company declared in its own filings?
The check stopped before any comparison: the program that reads company filings failed its hand check, so whether recorded payouts match declared amounts is not settled. Study 77's reading stands as published.
Study 78 · Changed a rule or plan · recorded 31 October 2026
Do speculative non-paying shares traded more than usual beat quieter ones overnight relative to the day, unlike paying shares?
Among the most speculative non-paying shares, the half busiest relative to usual ranked 0.0095 higher on a 0 to 1 scale (0.0017 to 0.0173) on next week's overnight less daytime return: 0.38 percentage points a week more (0.09 to 0.68).
Study 77 · Changed a rule or plan · recorded 30 October 2026
Does a share's gain at the open on its payout day rise with its payout yield?
On payout days the gain at the open rose with payout yield: 0.142 percentage points more in the highest-yield third than the lowest (0.072 to 0.212), and that third fell back from the open to the close.
Study 76 · Changed a rule or plan · recorded 30 October 2026
Does the clipped average keep its false-alarm rate when the compared groups' extreme values differ?
A proposal follows to stop clipped averages deciding comparisons of two groups: on ordinary days with no true difference, the clipped average reported a gap in 59% of tests against a 5% target where the groups' extreme values differed most.
Study 75 · Concluded · recorded 29 October 2026
After a large one-day fall that a company filing dates, do the overnight gaps carry a further decline of at least 0.5%?
Not at the size tested: in daily opening and closing prices, the further decline after these falls was measured in trading hours, while the two overnight gaps moved slightly upward relative to comparison shares.
Study 74 · No reading (stopped by its own checks) · recorded 29 October 2026
After a one-day fall that a company filing dates, does a third or more of the further decline happen overnight?
The question stays open: the study stopped before measuring any return, because a check fixed in advance found the source's adjusted and unadjusted closing prices out of step on some shares, so its two price constructions could not be compared.
Study 73 · Inconclusive · recorded 28 October 2026
Do smaller companies' shares gain a fifth of a point more against the benchmark fund before market holidays?
The question stays open: in 2018 to 2022, less the benchmark fund, smaller companies' shares returned 0.09 percentage points a session more on holiday eves than on their weekday's other sessions (-0.08 to 0.25), too uncertain to settle.
Study 72 · Changed a rule or plan · recorded 27 October 2026
Do the most speculative shares gain overnight and give it back during the trading day?
Speculative shares gained overnight and gave it back during the day: in 2018 to 2022, at a flat market, their overnight return less their daytime return beat the least speculative shares' by 0.77 percentage points a week (0.40 to 1.14).
Study 71 · Changed a rule or plan · recorded 26 October 2026
When a payout comes off a share, does its price fall by less than the payout?
The project stops looking for a payout shortfall in these shares: on the first day without the payout, 2018 to 2022, returns net of similar paying shares averaged -0.002 percentage points (-0.052 to 0.048), below the tenth set in advance.
Study 70 · No reading (stopped by its own checks) · recorded 26 October 2026
Among the least traded shares, do a week's winners still trail its losers when screened on the traded price?
Nothing changes yet: this check stopped before measuring any return, so the earlier finding and its price-screen limitation stand as published, and whether a corrected design is written is decided at the next review.
Study 69 · Inconclusive · recorded 25 October 2026
At a flat market, did shares the market fund explains most beat those it explains least over the next month?
Nothing changes: read at a flat market, the month's gap between the most and least explained fifths of shares in 2018 to 2022 was 0.557 points, with a range from -0.842 to 1.956, too wide to settle.
Study 68 · Changed a rule or plan · recorded 25 October 2026
Do smaller shares, taken together, do worse against the market fund in the month after a quarter ends than in the next two months?
No season adjustment is proposed: in 2018 to 2022, the month after a quarter ends differed from the next two months by 0.022 percentage points a day against the market fund, and a tenth of a point is ruled out.
Study 67 · Changed a rule or plan · recorded 24 October 2026
Matched on trading volume, do one-day falls of a tenth or more without a company filing still recover more?
Falls of a tenth or more without a company filing within a day still recovered more than falls with one at matched trading volume, by 0.80 points (0.25 to 1.35); the filings are grouped by their form next.
Study 66 · Changed a rule or plan · recorded 23 October 2026
Had companies whose large one-day fall came with a filing already trailed same-day fallers without one over the prior year?
Shares that fell sharply with a company filing within a day had not trailed same-day fallers without one by ten points over the prior year (a gap of -1.4 points, -5.5 to 3.0); quiet-day trading volume is compared next.
Study 65 · Changed a rule or plan · recorded 23 October 2026
Do large one-day rises with a company filing nearby keep rising from the next day's opening price?
From the next day's open, large rises with a filing nearby trailed quiet shares by 0.43 percentage points over three sessions rather than rising, while such falls kept falling; the programme asks next whether those fallers were already doing worse.
Study 64 · Changed a rule or plan · recorded 21 October 2026
After a two percent rise, do shares closing near the day's high beat those closing low from the next open?
After a big rise, shares closing near the day's high beat those closing in the lower half by 0.22 percentage points from the next open to that session's close. Our records propose allowing for the closing position in such studies.
Study 63 · Inconclusive · recorded 21 October 2026
Do speculative shares fare worse against safe shares from open to close after a weekend than before it?
No plan changes and no follow-up study is proposed: after weekends, speculative shares did 0.16 percentage points worse against safe shares from open to close than before, on average, but the weeks varied too much to settle it.
Study 62 · No reading (stopped by its own checks) · recorded 21 October 2026
After completing the purchase of another company, do buyers trail the average company in the second and third months?
The earlier verdict on these purchases stands and the line of work stays closed: the study stopped before measuring any buyer's returns, because it could not reliably detect a shortfall as small as the 2.5 percentage points fixed in advance.
Study 61 · Changed a rule or plan · recorded 20 October 2026
From the next opening price, do shares that fell with no company filing still beat those with one?
From the next day's opening price, shares that fell with no company filing still beat those with one by 1.5 percentage points over three sessions, because those with a filing kept falling; a planned recovery expectation is proposed for withdrawal.
Study 60 · Changed a rule or plan · recorded 19 October 2026
Do shares trail comparable companies after a late-report notice when the measure is checked for false alarms?
Re-tested on a measure whose false-alarm rate holds, the shortfall study 50 reported in the month after late-report notices is not established; a correction of study 50's answer goes to the owner as a proposal.
Study 59 · Inconclusive · recorded 19 October 2026
Do shares follow their closest-moving peers' past week after the weekend, but not between ordinary weekdays?
No peer-group control enters our event measures yet: after the weekend, shares whose closest-moving peers rose most beat those whose peers fell most by 0.19 points, a range from -0.06 to 0.43 that leaves the question open.
Study 58 · Changed a rule or plan · recorded 18 October 2026 · corrected
Among the least traded shares, do the week's biggest risers trail its biggest fallers the next week?
By the rule set in advance, event studies on thinly traded shares will control for the weekly reversal: among the least traded third, the week's top fifth trailed its bottom fifth by 0.51 points the next week.
Study 57 · Changed a rule or plan · recorded 16 October 2026 · corrected
Do shares that gained most overnight in one month keep gaining most overnight the next month?
The planned study of big one-day moves will measure from the opening price: shares in the top fifth by last month's overnight return beat the bottom fifth by 3.8 points overnight the next month, and gave it back by day.
Study 56 · Inconclusive · recorded 15 October 2026
Does the fallback our studies use, the average with extreme returns clipped, tilt their readings?
On ordinary months the clipped average's false-difference rate was 13% against a 5% target; whether it sat over the one-point tolerance below the plain average is not settled. On ordinary stretches it sat slightly lower, within tolerance, erring near target.
Study 55 · Changed a rule or plan · recorded 14 October 2026 · corrected
After a large one-day fall, do shares recover more over three sessions if no company filing explains it?
Falls of 10% or more that no company filing explained recovered 1.6 percentage points more over the next three sessions than explained falls on the same day; by the rule set in advance, a study of why comes next.
Study 54 · Inconclusive · recorded 14 October 2026
Do shares fall behind the market in the month before a regular payout cut takes effect?
No rule on the month before a payout cut is adopted: shares stayed about level with the market beyond their ordinary months, ruling out a fall of three points but not a gain of that size.
Study 53 · No reading (stopped by its own checks) · recorded 13 October 2026
After a company answers an outside bid for its shares, do they beat the market over two weeks?
The question stays open: the study stopped before measuring how any company's shares did after its answer to an outside bid, because its check fixed in advance found it could not reliably detect a drift of three percentage points.
Study 52 · Changed a rule or plan · recorded 13 October 2026
Does the year-long shortfall after a priced prospectus appear in drug and biotechnology companies and in other industries alike?
The shortfall after a priced prospectus held in drug and biotechnology companies and in other industries alike, so the earlier finding stands as general across industries. The two groups' shortfalls could not be told apart.
Study 51 · Changed a rule or plan · recorded 13 October 2026
Do one company's returns against comparable companies, at different dates, move together more than earlier studies assumed?
Two monthly returns of the same company, each against comparable companies, barely move together, within what earlier studies assumed; two yearly returns move together more, mostly because their years share trading days, so year-long plans will assume more.
Study 50 · Changed a rule or plan · recorded 12 October 2026 · corrected
Do shares fall behind comparable companies in the month after a company says its report will be late?
Study 50 reported shares trailing comparable companies in the month after a late-report notice, but a re-test on a measure whose false-alarm rate holds (study 60) did not establish it; a shortfall of about three points is not ruled out.
Study 49 · Changed a rule or plan · recorded 11 October 2026
Over the year after a company raises its regular payout, do its shares pull ahead of comparable companies?
No year-long expectation of a lead after a raise of the regular payout is adopted: in approximate average wealth, the registered measure, payers' shares ended the year level with comparable companies, ruling out a gap of the size tested.
Study 48 · Changed a rule or plan · recorded 11 October 2026
Do low-priced shares fare differently from comparable companies in the year after a large one-day fall?
After a large one-day fall, the typical holder's wealth in low-priced shares ended the year about 17.5% below comparable companies', a shortfall that did not persist under the volatility adjustment fixed in advance, so no year-long expectation is adopted.
Study 47 · Changed a rule or plan · recorded 10 October 2026
Do a company's shares trail comparable companies over the month and the year after a priced prospectus?
Shares of companies filing a priced prospectus trailed comparable companies over the month and year after the sale, and the year's shortfall was measurably smaller where shares had risen more. These sales become a family of events for further study.
Study 46 · Changed a rule or plan · recorded 10 October 2026
Can the earlier priced-prospectus study's one-year test be made able to detect the effect size set in advance?
The earlier study's one-year test can be made precise enough on the same sales: large rises by a small number of selling companies in the year before the sale set its spread, and a logarithmic scale brings it within target.
Study 45 · No reading (stopped by its own checks) · recorded 10 October 2026
After a company files a priced prospectus, do its shares trail comparable companies afterwards?
The question stays open: the study stopped before measuring how any company's shares did after a priced prospectus, because over the year it could not reliably detect a shortfall as small as the 10 percentage points fixed in advance.
Study 44 · Changed a rule or plan · recorded 8 October 2026
Can matching by company name tie the shares a data vendor's list left out to their filings?
Questions about company filings can now cover the whole development sample: the vendor's full stored list ties nearly every share to its filings, and matching by company name passed its accuracy check and stays as a cross-check.
Study 43 · No reading (stopped by its own checks) · recorded 8 October 2026
Over the year after a payout cut, does a company trail similar companies more than after a payout raise?
The question stays unanswered and the earlier findings on payout changes stand: fewer payout cuts qualified than the minimum the study fixed in advance, so it stopped before measuring how any company did afterwards.
Study 42 · Checked a method · recorded 7 October 2026
Does our bootstrap test keep its false-alarm rate on made-up bell-curve values at other sample sizes?
The bootstrap test's soundness check now covers four sample sizes, not one: on made-up bell-curve values it kept its false-alarm rate, so an earlier one-sided excess belongs to real returns after share splits, not to the test.
Study 41 · Inconclusive · recorded 6 October 2026
Do spun-off companies trail the market in their first trading days, and does any shortfall last?
This line of work closes, with no sharper design proposed. The result is not settled: for spun-off companies neither a recovering selling dip nor a lasting markdown was established, and a modest early shortfall is not ruled out.
Study 40 · No reading (stopped by its own checks) · recorded 6 October 2026
Do three earlier stock-split findings hold against a yardstick whose false-alarm rate stays on target?
Three earlier stock-split findings keep the status they had: the re-check could not be made, because no yardstick tried kept its false-alarm rate within the bounds set in advance on made-up data, so no real data was read.
Study 39 · Inconclusive · recorded 5 October 2026
After buying another company, does a buyer trail similar companies more when its shares had risen most beforehand?
No plan changed and work on these deals rests, because the result is inconclusive: neither a buyer's earlier share-price rise nor the deal itself was shown to explain how its shares did in the year after it bought another company.
Study 38 · Changed a rule or plan · recorded 4 October 2026
Against a recent level, do wide gaps between paired shares close more often than chance?
Work on paired shares reopens, but only as a finding on development data: judged against a recent level, wide price gaps closed a little more often than chance, though short-lived swings in daily moves could explain it.
Study 37 · Changed a rule or plan · recorded 4 October 2026
Do three checks for planted data mistakes still hold in other settings?
No data check joins the shared checks without its limits written beside it: three checks for planted mistakes each held in some settings and fell short in others, so a study relying on one must name what it misses.
Study 36 · Inconclusive · recorded 3 October 2026
Does the month after a payout change follow the change's size or earlier returns?
No trading rule built on payout changes takes a change's size or the company's earlier returns into account, because the result is inconclusive: neither was shown to shape the month after the change.
Study 35 · Changed a rule or plan · recorded 3 October 2026
After a steep one-day fall, does a low-priced share rebound overnight beyond its bid-ask spread?
The overnight part of an earlier rebound in low-priced shares is now recorded as a quoting effect, and that line of work rests: the rebound fit within the bid-ask spread and the share's usual overnight move.
Study 34 · Inconclusive · recorded 2 October 2026 · corrected
After a large one-day move, do rises and falls differ in returns against quiet peers?
Nothing changes, because the result is inconclusive: after a large one-day move, shares trailed quiet peers of the same volatility, but whether by the size fixed as meaningful is undecided; rises and falls could not be told apart.
Study 33 · Changed a rule or plan · recorded 2 October 2026 · corrected
Does trimming extreme values restore a bootstrap test's false-alarm rate on heavy-tailed returns?
A planning rule changed: the bootstrap test now applies only where its own false-alarm check passes. Removing at most 2% of the most extreme values restored its false-alarm rate to target on the lighter-tailed of two pools, not the heavier.
Study 32 · Concluded · recorded 2 October 2026 · corrected
Does a low-priced share's rebound after a steep one-day fall still appear when measured from the next day's opening price?
No rule or plan changed: from the next day's open, the rebound after steep one-day falls in low-priced shares was gone overall, undecided in the smallest third of companies, and could not be told from an effect of quoted prices.
Study 31 · Changed a rule or plan · recorded 1 October 2026 · corrected
Can simple automatic checks catch three planted data defects without flagging clean copies?
Planning rules changed: checks for duplicated rows and for a subset chosen after looking caught every planted copy without false alarms and are proposed for building, while no check caught a single extreme row cleanly.
Study 30 · Inconclusive · recorded 1 October 2026 · corrected
Does a company separated from its parent beat comparable companies over the following year?
No follow-up is planned, because the result is inconclusive: the one-year returns of companies newly separated from their parents could not be told apart from those of comparable companies.
Study 29 · Inconclusive · recorded 1 October 2026 · corrected
After a first cash payout, do next-month returns differ between recurring and one-off payouts?
No follow-up on payout type is planned, because the result is inconclusive: in the month after a payout that ended a year without one, returns after recurring and one-off payouts could not be told apart.
Study 28 · Not supported · recorded 1 October 2026 · corrected
Do moves flagged by only one of two rules lead to returns beyond the market?
The monitoring rule stays as it is: moves flagged by only one of two rules, the current one or a market-wide alternative, were not followed by returns beyond the market as large as the size fixed in advance.
Study 27 · Changed a rule or plan · recorded 1 October 2026 · corrected
Does our standard test's lopsided false-alarm rate depend on skew rather than heavy tails?
Planning rules changed: our standard test's lopsided false alarms are now treated as a property of skew, and its bootstrap remedy stays limited to the stock split studies after failing on heavier tails.
Study 26 · Changed a rule or plan · recorded 1 October 2026 · corrected
Can a data vendor's filer numbers tie at least 85% of our shares to their old company filings?
Planning changed: questions about company filings will not use a data vendor's filer numbers, which tied only 51.5% of the shares we study to their filings; a filing question first needs another way to match them.
Study 25 · Inconclusive · recorded 1 October 2026 · corrected
Do price gaps between shares that move together close faster than a reshuffle of the same moves?
Nothing changed: the study led to no further research on pairs of shares that move together. How often their price gaps closed could not be told apart from a reshuffle of the same moves, though we expected faster closing.
Study 24 · Inconclusive · recorded 1 October 2026 · corrected
Do shares return less in the month after a regular payout cut than in the same company's ordinary payout months?
Nothing changed: payout cuts were not carried to a planned study. Returns in the month after a regular cut leaned lower than in the same company's ordinary months, but within the range chance produces, so the answer is not settled.
Study 23 · Changed a rule or plan · recorded 30 September 2026 · corrected
Study 23
Study 22 · Inconclusive · recorded 30 September 2026 · corrected
Study 22
Study 21 · Changed a rule or plan · recorded 30 September 2026 · corrected
Are there enough pairs of shares that move together to study?
Study 20 · Changed a rule or plan · recorded 30 September 2026 · corrected
Study 20
Study 19 · Inconclusive · recorded 30 September 2026 · corrected
Study 19
Study 18 · Concluded · recorded 30 September 2026 · corrected
Do cheap stocks bounce back after a bad day?
Study 17 · Checked a method · recorded 30 September 2026 · corrected
Does our checking method agree with published findings?
Study 16 · Changed a rule or plan · recorded 29 September 2026 · corrected
Study 16
Study 15 · Changed a rule or plan · recorded 29 September 2026 · corrected
Study 15
Study 14 · Changed a rule or plan · recorded 28 September 2026 · corrected
Study 14
Study 13 · Changed a rule or plan · recorded 28 September 2026 · corrected
Study 13
Study 12 · Changed a rule or plan · recorded 28 September 2026 · corrected
Study 12
Study 11 · Changed a rule or plan · recorded 27 September 2026 · corrected
Study 11
Study 10 · No reading (stopped by its own checks) · recorded 27 September 2026 · corrected
Do one-week tests hold their false-alarm rate at longer windows?
Study 9 · Checked a method · recorded 27 September 2026 · corrected
Do shares still return less in the five sessions after a split that adds shares than in their ordinary sessions?
Study 8 · Changed a rule or plan · recorded 27 September 2026 · corrected
Do the number of random draws and the rule for ties change our tests' false-alarm and detection rates?
Study 7 · Changed a rule or plan · recorded 27 September 2026 · corrected
How many split events that increase the number of shares would a future test need?
Study 6 · Changed a rule or plan · recorded 27 September 2026 · corrected
Why does a test on average return differences raise false alarms more often on one side than the other?
Study 5 · Changed a rule or plan · recorded 27 September 2026 · corrected
Does a rank-based test detect a planted fall more often than a test on averages when fewer companies carry it?
Study 4 · Changed a rule or plan · recorded 26 September 2026 · corrected
Why did a rank-based test raise false alarms in 5.6% of no-effect trials on share-adding splits, against its 5% target?
Study 3 · No reading (stopped by its own checks) · recorded 26 September 2026 · corrected
Does the gap in five-session returns after reverse splits also appear when ordinary sessions replace the splits?
Study 2 · Changed a rule or plan · recorded 26 September 2026 · corrected
Does our bootstrap test keep its 5% false-alarm rate on stock split data with ten times the simulations?
Planning rules changed: future stock split studies with fewer than 548 events must set their false-alarm thresholds by simulation, because the bootstrap test raised too many false alarms at smaller sizes in a check with ten times the simulations.
Study 1 · Inconclusive · recorded 26 September 2026 · corrected
Is our standard test's lopsided false-alarm rate on stock splits a general effect of skew?
Nothing changed: the planning rules for stock split studies stay limited to them. Shares that never split showed the same lopsided false alarms, but the check meant to tell skew from heavy tails failed, so the cause is not settled.