A thread is one line of inquiry: a first hypothesis and everything the lab learned from it. Each test produces observations; an investigator reasons about them, asks a sharper question, and proposes the next hypothesis to answer it. Open a thread to read it like a lab notebook, most recent work first here.
Loading threads…
117 threads, most recently active first. 322 archive imports that never grew are hidden. Show them.
PROMISINGstory39 hypotheses, 12 deep23 open questions3 observations waitingH36-pulled-up-after-being-held-up
best rung: Live
Observed: H-f956317c (Test-mechanism: big-field heavy beatings where the horse was never inv) survived falsification: 5 of 6 attacks left it standing; it is frozen for its one vault query.
PROMISINGtrajectory19 hypotheses, 5 deep10 open questions5 observations waitingH85-fell-when-going-well
best rung: Live
Observed: H-6e4258ac (Test-mechanism: chase unseaters that were well fancied last time beat ) met the sealed vault supportive: A/E None on 0 bets (p 1.0, threshold 0.0019).
SHADOWshock4 hypotheses, 1 deep3 open questionsS2-chase-completed-normal-habitual-poor-jumper
best rung: Shadow
Asked: Is the holdout lift caused by the poor-jumping condition (e), or would any horse that completed a chase last time to expectation (s normal) show it?
PROMISINGseason-ireland10 hypotheses, 2 deep3 open questionsH228-dual-purpose-horse-in-an-irish-summer-flat-handicap
best rung: Live
Verdict: fails: 256 bets, needs 300; A/E 0.981, needs 0.95 or lower; p 0.5276 (lower tail), needs 0.05; periods [0.98], each needs 0.99 or lower; A/E 0.981 against same-race control 1.011, needs 0.03 below
SHADOWshock3 hypotheses, 1 deep2 open questionsS3-chase-not-completed-large-negative
best rung: Shadow
Asked: Among last-time chase non-finishers, does a deeper jumping shock (more negative s) lead to a higher A/E next time, or is s irrelevant once the horse failed to finish?
PROMISINGstory27 hypotheses, 5 deep20 open questionsH24-one-off-jumping-error-well-beaten
best rung: Live
Answered: No: chase pull-ups with clean jumping that led last time score 0.68 on 43 (CI 0.14 to 1.97), the same as the whole clean pool (0.71 on 208). The broad led/prominent pull-up rival (H-a2870576) was killed, so run style does not concentrate the under-performance.
KILLEDnegative10 hypotheses, 3 deep8 open questionsH-220ad315
best rung: Shadow
Answered: No. Quick-returning winners whose last win came at 7/1+ scored 0.94 on 228, near par and much shallower than the 0.62 parent, so the shortfall is not concentrated in surprise winners.
KILLEDnegative5 hypotheses, 1 deep3 open questionsH-bfcb9669
best rung: Discovery
Answered: No. Poor hurdle jumpers stepping up to fences scored 0.83 on 150, no stronger than the parent's 0.78. The deficit sits entirely in Ireland (0.44 on 57), while GB was at par (1.03 on 93).
KILLEDnegative2 hypotheses, 1 deep2 open questions1 observations waitingH-dd2d6d12
best rung: Discovery
Observed: H-07082d28 (Winners gifted an uncontested lead are over-bet at the front of the ma) pointed the predicted way but failed the bar: A/E 0.93 on 45 bets (p 0.64), control 1.03.
PROMISINGstory18 hypotheses, 5 deep4 open questionsH34-won-easily-last-time
best rung: Live
Answered: No. Big-rise quick-return winners stepping up 15%+ in trip scored 0.67 on 50 (p 0.879, CI 0.25 to 1.47), and only 5 of them were 16/1+. So a trip step-up cannot explain the 0-from-32 longshot failures.
PROMISINGrace-betting12 hypotheses, 4 deep15 open questionsH199-favourite-at-a-big-price-in-a-big-handicap
best rung: Live
Verdict: fails: A/E 0.981, needs 1.08; p 0.5685, needs 0.05 or lower; periods [1.18, 0.71], each needs 1.0; A/E 0.981 against same-race control 1.03, needs 0.03 above (not a niche pass: A/E 0.981 below 1.08; p 0.568478005324285 above 0.5; does not beat the same-race control 1.03; a period points the other wa
PROMISINGnamed-trainer6 hypotheses, 2 deep5 open questionsH17-jamie-snowden-novices
best rung: Live
Answered: No. Over jumps the lift is novice-specific: Snowden's non-novice, non-maiden hurdlers and chasers scored 0.975 on 187 (periods 1.04 and 0.84, control 1.01), so 'novice' is not just a label on a general trainer effect.
PROMISINGstory11 hypotheses, 1 deep7 open questionsH19-momentum-interrupted-beaten-narrowly
best rung: Live
Answered: The data can't answer this yet. Checked-and-recovered runners scored 1.59 on 31 bets (control 0.87) but were killed when 1 of 1 attacks broke them, with too few bets per period and country, so they can't be told apart from the parent.
PROMISINGage-stamina20 hypotheses, 4 deep16 open questionsH175-young-horse-in-a-marathon-chase
best rung: Live
Answered: Not convincingly. Young chasers with 6 or fewer chase runs at 3m to 3m3f scored 0.93 on 743 bets (CI 0.76 to 1.13, p 0.78) against a same-race control of 1.06, so the sign leans negative but is not distinguishable from par, and 0 wins from 107 bets at 16/1+ carry much of it.
PROMISINGseasonal32 hypotheses, 5 deep31 open questionsH126-lightly-raced-four-year-old-returns
best rung: Live
Answered: No. The year-round turf-to-AW returner rule after 150+ days scored 1.14 at discovery but 0.979 on 3155 in the sealed vault (p 0.64), so there is no robust year-round AW switch effect.
PROMISINGletters12 hypotheses, 4 deep9 open questionsH98-brought-down-or-unseated
best rung: Live
Verdict: fails: A/E 0.994, needs 1.05; p 0.6058, needs 0.05 or lower; periods [1.01, 0.96], each needs 1.0; A/E 0.994 against same-race control 0.995, needs 0.03 above (not a niche pass: A/E 0.994 below 1.05; p 0.6058069056779608 above 0.5; does not beat the same-race control 0.995)
PROMISINGtrack-stamina15 hypotheses, 5 deep9 open questionsH181-proven-over-further-at-a-stiff-track
best rung: Live
Asked: When the frozen stiff-sprint rules (H-2b483ac5, H-cbd60237) build up forward bets, does the A/E hold near 1.2 in GB, and does it stay near 1.0 or below in Ireland?
PROMISINGage5 hypotheses, 1 deep4 open questionsH45-young-horses-in-novice-chases
best rung: Live
Verdict: fails: A/E 0.974, needs 1.05; p 0.6117, needs 0.05 or lower; periods [1.0, 0.88], each needs 1.0; A/E 0.974 against same-race control 1.16, needs 0.03 above (not a niche pass: A/E 0.974 below 1.05; p 0.6116703107013322 above 0.5; does not beat the same-race control 1.16; a period points the other wa
Observed: IQ111-control-second-try-at-3m-after-finishing-the-first (IQ111: control second try at 3m after finishing the first) was GREEN in the earlier research: A/E 1.06 on 3098 bets (periods 1.06 / 1.08 / 1.05); now KILLED.
Observed: H152-twiston-davies-at-cheltenham-before-christmas (H152: twiston davies at cheltenham before christmas) was predicted positive but went the other way in the earlier research: A/E 0.91 on 122 bets (periods 0.99 / 1.05).
Observed: IQ35-lost-ground-everywhere (IQ35: lost ground everywhere) was GREEN in the earlier research: A/E 1.04 on 7180 bets (periods 1.03 / 1.05); now PROMISING.
Observed: IQ141c-wide-early-recovered-no (IQ141c: wide early recovered no) was GREEN in the earlier research: A/E 1.62 on 159 bets (periods 1.37 / 1.80); now PROMISING.
Observed: IQ113-control-second-try-at-3m-after-passing-the-first (IQ113: control second try at 3m after passing the first) was GREEN in the earlier research: A/E 1.06 on 333 bets (periods 1.05 / 1.08); now KILLED.
Observed: S4-chase-completed-large-positive (S4: chase completed large positive) pointed the predicted way (positive) on the sealed holdout but too weakly to prove it: A/E 1.174 on 156 bets (p 0.218, false-discovery threshold 0.0034); tracked forward in SHADOW.
Observed: IQ61b-pulled-up-jumped-well-then-pulled-up-all prices (IQ61b: pulled up jumped well then pulled up all prices) was GREEN in the earlier research: A/E 1.38 on 362 bets (periods 1.16 / 1.99); now PROMISING.
Observed: H109-the-handicap-plot (H109: the handicap plot) was GREEN in the earlier research: A/E 1.05 on 4608 bets (periods 1.07 / 1.02 / 1.10); now PROMISING.
Observed: IQ10-went-fast-early-jumped-well-and-faded (IQ10: went fast early jumped well and faded) was GREEN in the earlier research: A/E 1.06 on 574 bets (periods 1.08 / 1.05); now PROMISING.
Observed: IQ143c-one-off-jumping-error-no (IQ143c: one off jumping error no) was GREEN in the earlier research: A/E 1.88 on 161 bets (periods 1.45 / 2.40); now PROMISING.
Observed: IQ112-second-try-at-3m-after-a-visible-failure (IQ112: second try at 3m after a visible failure) was predicted positive but went the other way in the earlier research: A/E 0.80 on 172 bets (periods 0.80 / 0.80).
Observed: IQ37-lost-ground-jumping-without-going-fast (IQ37: lost ground jumping without going fast) was GREEN in the earlier research: A/E 1.06 on 7438 bets (periods 1.02 / 1.16); now PROMISING.
Observed: IQ53-failed-to-finish-jumping-out-of-character (IQ53: failed to finish jumping out of character) was predicted negative but went the other way on the sealed holdout: A/E 1.11 on 1095 bets.
Observed: H150-plot-yards-in-handicaps-after-a-break (H150: plot yards in handicaps after a break) was predicted positive but went the other way in the earlier research: A/E 0.71 on 257 bets (periods 0.76 / 0.57 / 1.20).
Observed: H118-out-of-the-race-early (H118: out of the race early) was predicted positive but went the other way in the earlier research: A/E 0.87 on 281 bets (periods 0.83 / 0.95).
Observed: H23-hurdle-prep-into-a-chase (H23: hurdle prep into a chase) was GREEN in the earlier research: A/E 1.14 on 3708 bets (periods 1.13 / 1.12 / 1.18); now PROMISING.
Observed: H27-flat-run-into-a-hurdle (H27: flat run into a hurdle) was GREEN in the earlier research: A/E 1.07 on 3339 bets (periods 1.18 / 1.00 / 1.04); now PROMISING.
Observed: IQ18-chaser-whose-jumping-is-improving (IQ18: chaser whose jumping is improving) was GREEN in the earlier research: A/E 1.05 on 2252 bets (periods 1.10 / 1.00); now PROMISING.
Observed: IQ65b-pulled-up-poor-jumper-anyway-all prices (IQ65b: pulled up poor jumper anyway all prices) was GREEN in the earlier research: A/E 1.80 on 207 bets (periods 1.88 / 1.67); now PROMISING.
Observed: IQ12-made-up-places-mid-race-while-losing-ground-jumping (IQ12: made up places mid race while losing ground jumping) was GREEN in the earlier research: A/E 0.90 on 653 bets (periods 0.94 / 0.86); now PROMISING.
Observed: IQ151c-no-error-large-cost (IQ151c: no error large cost) was GREEN in the earlier research: A/E 1.14 on 989 bets (periods 1.10 / 1.27); now PROMISING.
Observed: TD2-durrell-harry-elsewhere (TD2: durrell harry elsewhere) was GREEN in the earlier research: A/E 1.23 on 370 bets (periods 1.01 / 1.54); now PROMISING.
Observed: H120-one-reason-to-forgive-no-negative (H120: one reason to forgive no negative) was GREEN in the earlier research: A/E 1.17 on 5189 bets (periods 1.12 / 1.23); now PROMISING.
Observed: H104-yard-good-at-chase-debuts (H104: yard good at chase debuts) was GREEN in the earlier research: A/E 1.15 on 791 bets (periods 1.11 / 1.12 / 1.33); now PROMISING.
Observed: H15-yard-good-with-recruits (H15: yard good with recruits) was GREEN in the earlier research: A/E 1.08 on 4195 bets (periods 0.99 / 1.11 / 1.13); now PROMISING.
PROMISINGpulled-up-overlay1 hypothesis1 observations waitingIQ61a-pulled-up-jumped-well-then-pulled-up-system, 14-1 or shorter
best rung: Promising
Observed: IQ61a-pulled-up-jumped-well-then-pulled-up-system, 14-1 or shorter (IQ61a: pulled up jumped well then pulled up system, 14 1 or shorter) was GREEN in the earlier research: A/E 1.35 on 156 bets (periods 1.10 / 2.08); now PROMISING.
Observed: S8-hurdle-completed-large-positive (S8: hurdle completed large positive) was predicted positive but went the other way on the sealed holdout: A/E 0.86 on 408 bets.
PROMISINGpulled-up-overlay1 hypothesis1 observations waitingIQ60a-pulled-up-has-figures-system, 14-1 or shorter
best rung: Promising
Observed: IQ60a-pulled-up-has-figures-system, 14-1 or shorter (IQ60a: pulled up has figures system, 14 1 or shorter) was GREEN in the earlier research: A/E 1.32 on 481 bets (periods 1.26 / 1.50); now PROMISING.
Observed: IQ33-costly-jumping-with-nothing-in-the-remark (IQ33: costly jumping with nothing in the remark) was GREEN in the earlier research: A/E 1.09 on 3604 bets (periods 1.11 / 1.03); now PROMISING.
Observed: H195-flat-speed-over-hurdles-on-winter-good-ground (H195: flat speed over hurdles on winter good ground) was GREEN in the earlier research: A/E 1.16 on 478 bets (periods 1.27 / 1.19 / 0.76); now PROMISING.
Observed: H84-in-front-too-far-from-home (H84: in front too far from home) was GREEN in the earlier research: A/E 1.03 on 1348 bets (periods 0.97 / 1.07); now PROMISING.
Observed: IQ141a-wide-early-recovered-with-figures (IQ141a: wide early recovered with figures) was GREEN in the earlier research: A/E 1.45 on 253 bets (periods 1.33 / 1.54); now PROMISING.
Observed: S5-hurdle-completed-large-negative (S5: hurdle completed large negative) pointed the predicted way (positive) on the sealed holdout but too weakly to prove it: A/E 1.203 on 239 bets (p 0.179, false-discovery threshold 0.0034); tracked forward in SHADOW.
Observed: H190-seasoned-chaser-at-stiff-fences (H190: seasoned chaser at stiff fences) was GREEN in the earlier research: A/E 1.05 on 2968 bets (periods 1.03 / 1.05 / 1.09); now PROMISING.
Observed: IQ32-mistake-in-the-remark-that-cost-ground (IQ32: mistake in the remark that cost ground) was GREEN in the earlier research: A/E 1.08 on 1801 bets (periods 1.03 / 1.22); now PROMISING.
Observed: VET8-younger-heavy-defeat-engine-intact (VET8: younger heavy defeat engine intact) was GREEN in the earlier research: A/E 1.30 on 219 bets (periods 1.24 / 1.42); now PROMISING.
Observed: H87-lost-ground-through-the-race (H87: lost ground through the race) was GREEN in the earlier research: A/E 1.07 on 2247 bets (periods 1.06 / 1.08); now PROMISING.
Observed: HS4-harry-on-dan-usually-prominent (HS4: harry on dan usually prominent) was predicted positive but went the other way in the earlier research: A/E 0.77 on 184 bets (periods 0.63 / 0.91).
Observed: H140-mullins-hurdlers-in-flat-staying-races (H140: mullins hurdlers in flat staying races) was predicted positive but went the other way in the earlier research: A/E 0.91 on 239 bets (periods 0.85 / 0.83 / 1.23).
Observed: AB2-veterans-aged-11 (AB2: veterans aged 11) was predicted positive but went the other way in the earlier research: A/E 0.89 on 419 bets (periods 0.81 / 1.01 / 0.84).
Observed: HS2-harry-on-dan-held-up-last-time (HS2: harry on dan held up last time) was GREEN in the earlier research: A/E 1.08 on 186 bets (periods 1.10 / 1.06); now PROMISING.
Observed: H20-wide-early-and-got-back-into-it (H20: wide early and got back into it) was GREEN in the earlier research: A/E 1.43 on 313 bets (periods 1.28 / 1.57); now PROMISING.
Observed: H97-yard-in-form-and-fresh (H97: yard in form and fresh) was GREEN in the earlier research: A/E 1.11 on 3334 bets (periods 1.13 / 1.08 / 1.17); now PROMISING.
Observed: H229-favourite-at-a-big-public-meeting (H229: favourite at a big public meeting) was GREEN in the earlier research: A/E 1.05 on 2065 bets (periods 1.03 / 1.06 / 1.08); now PROMISING.
Observed: IQ34-lost-ground-in-one-place (IQ34: lost ground in one place) was GREEN in the earlier research: A/E 1.07 on 479 bets (periods 1.01 / 1.22); now PROMISING.
Observed: H221-busy-horse-on-the-winter-all-weather (H221: busy horse on the winter all weather) was predicted positive but went the other way in the earlier research: A/E 0.89 on 2059 bets (periods 0.80 / 0.90 / 1.02).
Observed: H50-wide-trip-by-track-geometry (H50: wide trip by track geometry) was GREEN in the earlier research: A/E 1.20 on 1288 bets (periods 1.11 / 1.27); now PROMISING.
Observed: IQ58-failed-to-finish-poor-jumper-bad-day (IQ58: failed to finish poor jumper bad day) was predicted positive but went the other way on the sealed holdout: A/E 0.79 on 505 bets.
Observed: IQ110-second-try-at-3m-after-not-finishing-the-first (IQ110: second try at 3m after not finishing the first) was GREEN in the earlier research: A/E 1.20 on 868 bets (periods 1.38 / 1.02 / 1.03); now PROMISING.
Observed: H60-second-season-chaser-fresh (H60: second season chaser fresh) was GREEN in the earlier research: A/E 1.11 on 1402 bets (periods 1.19 / 1.05 / 1.08); now PROMISING.
Observed: S1-chase-completed-large-negative (S1: chase completed large negative) pointed the predicted way (positive) on the sealed holdout but too weakly to prove it: A/E 1.001 on 84 bets (p 0.546, false-discovery threshold 0.0034); tracked forward in SHADOW.
Observed: IQ143a-one-off-jumping-error-with-figures (IQ143a: one off jumping error with figures) was GREEN in the earlier research: A/E 1.50 on 340 bets (periods 1.25 / 1.90); now PROMISING.
Observed: S7-hurdle-not-completed-large-negative (S7: hurdle not completed large negative) was predicted positive but went the other way on the sealed holdout: A/E 0.71 on 273 bets.
Observed: VET7r-old-heavy-defeat-declined (VET7r: old heavy defeat declined) was predicted positive but went the other way in the earlier research: A/E 0.83 on 186 bets (periods 0.95 / 0.52).
Observed: IQ19-chaser-whose-jumping-is-getting-worse (IQ19: chaser whose jumping is getting worse) was GREEN in the earlier research: A/E 1.07 on 2456 bets (periods 1.02 / 1.11); now PROMISING.
Observed: IQ63b-pulled-up-out-of-character-jumping-all prices (IQ63b: pulled up out of character jumping all prices) was predicted positive but went the other way in the earlier research: A/E 0.56 on 115 bets (periods 0.32 / 1.16).
Observed: H116-horse-that-needs-a-run (H116: horse that needs a run) was GREEN in the earlier research: A/E 1.05 on 5752 bets (periods 0.26 / 1.02 / 1.14); now PROMISING.
Observed: IQ20-one-bad-jumping-day-from-a-good-jumper (IQ20: one bad jumping day from a good jumper) was GREEN in the earlier research: A/E 1.08 on 732 bets (periods 1.00 / 1.13); now PROMISING.
Observed: VET5-younger-up-in-trip (VET5: younger up in trip) was GREEN in the earlier research: A/E 1.05 on 1918 bets (periods 1.07 / 1.01); now PROMISING.
Observed: IQ02-jumped-well-in-an-ugly-result (IQ02: jumped well in an ugly result) was GREEN in the earlier research: A/E 1.03 on 2270 bets (periods 1.00 / 1.07); now PROMISING.
Observed: H207-top-weight-in-a-wide-handicap (H207: top weight in a wide handicap) was predicted positive but went the other way in the earlier research: A/E 0.91 on 2381 bets (periods 0.82 / 1.06 / 0.91).
Observed: IQ82-jumps-folds-when-challenged (IQ82: jumps folds when challenged) was predicted negative but went the other way in the earlier research: A/E 1.09 on 552 bets (periods 1.25 / 0.92).
Observed: IQ64b-pulled-up-jumping-as-usual-all prices (IQ64b: pulled up jumping as usual all prices) was GREEN in the earlier research: A/E 1.36 on 267 bets (periods 1.29 / 1.53); now PROMISING.
Observed: H209-top-weight-in-a-claimer (H209: top weight in a claimer) was predicted positive but went the other way in the earlier research: A/E 0.91 on 800 bets (periods 0.86 / 0.93 / 1.10).
Observed: H12-yard-good-at-handicap-debuts (H12: yard good at handicap debuts) was GREEN in the earlier research: A/E 1.09 on 4046 bets (periods 1.13 / 1.00 / 1.34); now PROMISING.
Observed: H121-handicap-debut-from-a-good-yard-in-form (H121: handicap debut from a good yard in form) was GREEN in the earlier research: A/E 1.16 on 1811 bets (periods 1.27 / 1.02 / 1.52); now PROMISING.
Observed: H124-fit-hurdler-making-its-chase-debut (H124: fit hurdler making its chase debut) was predicted positive but went the other way in the earlier research: A/E 0.91 on 1173 bets (periods 1.04 / 0.81 / 0.87).
Observed: H134-chase-debut-from-a-good-yard-in-form (H134: chase debut from a good yard in form) was GREEN in the earlier research: A/E 1.12 on 365 bets (periods 1.17 / 1.10 / 1.06); now PROMISING.
Observed: H183-first-try-at-a-longer-trip-in-the-mud-at-a-stiff-track (H183: first try at a longer trip in the mud at a stiff track) was GREEN in the earlier research: A/E 1.14 on 2858 bets (periods 1.09 / 1.18 / 1.08); now PROMISING.
Observed: H149-charlie-appleby-short-prices-in-britain (H149: charlie appleby short prices in britain) was GREEN in the earlier research: A/E 1.05 on 715 bets (periods 1.01 / 1.09 / 1.07); now PROMISING.
Observed: H11-yard-good-fresh (H11: yard good fresh) was GREEN in the earlier research: A/E 1.07 on 7573 bets (periods 1.07 / 1.04 / 1.15); now PROMISING.
Observed: H198-been-round-the-national-fences-before (H198: been round the national fences before) was predicted positive but went the other way in the earlier research: A/E 0.84 on 334 bets (periods 1.05 / 0.64 / 0.90).
Verdict: fails: 28 bets, needs 150; A/E 0.998, needs 1.1; p 0.5784, needs 0.05 or lower; periods [], each needs 1.0; A/E 0.998 against same-race control 0.996, needs 0.03 above (not a niche pass: only 28 bets; A/E 0.998 below 1.1; p 0.5783763804363621 above 0.5; does not beat the same-race control 0.996)
Verdict: fails: A/E 0.875, needs 1.05; p 0.9393, needs 0.05 or lower; periods [0.8, 0.98], each needs 1.0; A/E 0.875 against same-race control 1.017, needs 0.03 above (not a niche pass: A/E 0.875 below 1.05; p 0.9393122723385944 above 0.5; does not beat the same-race control 1.017; a period points the other
Verdict: fails: A/E 1.019, needs 1.08; p 0.3269, needs 0.05 or lower; periods [1.06, 0.98], each needs 1.0; A/E 1.019 against same-race control 0.996, needs 0.03 above (not a niche pass: A/E 1.019 below 1.08; does not beat the same-race control 0.996)
Verdict: fails: A/E 0.942, needs 1.05; p 0.8668, needs 0.05 or lower; periods [0.84, 1.0], each needs 1.0; A/E 0.942 against same-race control 1.01, needs 0.03 above (not a niche pass: A/E 0.942 below 1.05; p 0.8668323334758147 above 0.5; does not beat the same-race control 1.01; a period points the other wa
Verdict: fails: A/E 0.99, needs 1.08; p 0.5476, needs 0.05 or lower; periods [0.86, 1.06], each needs 1.0; A/E 0.99 against same-race control 0.998, needs 0.03 above (not a niche pass: A/E 0.99 below 1.08; p 0.5475723737607818 above 0.5; does not beat the same-race control 0.998; a period points the other wa