The 19 August Research Instrument

← Elite Pass · Ledger opened 19 August 2026 · 1 day settled

This ledger records the original prospective research instrument, opened 19 August 2026. Current Elite Pass Race Intelligence is recorded separately, through the ongoing Daily prospective record. The two are kept deliberately apart: they are different instruments with different definitions, and merging them would make one look older than it is. That is why dates elsewhere on Elite Pass post-date this page. See today’s Race Intelligence →

Every research race day held here was recorded and frozen before the first relevant race, and settled against the exchange the same evening. No tips are sold. No stakes are placed. The record exists so that any claim made later can be checked against something that was fixed beforehand.

19 August 2026

FREEZE & SETTLEMENT

Recorded before the first race.

Frozen at10:39 BST

First race13:50

Races recorded36

Runners recorded407

Outcome data accessedNo

VerificationPre-race record sealed

Verify the technical record

File SHA-256: this is the one you can check

c8a54ce9 e6400743 388541e9 557606dc
db34a418 372026d1 7bbb54e6 2da35cbe

Content digest recorded at freeze, inside the file

f2a6f869 460b85ee de85059f 26b81d57
0ce0e058 840e09d8 9a32ce9d 96a8a59b

Source file

e7c7e837

How we know this was not written afterwards

This race day was recorded in full before the first race ran, and sealed with the digest shown above. The digest is taken from the original record, so if a single figure had been changed after the racing it would no longer match.

Technical verification

The original pre-race record is retained unaltered in the research archive and is available to a researcher or journalist on request. Its SHA-256 file digest is published above and reproduces byte for byte.

The second digest is the instrument’s own content seal, written into the record before racing. A file cannot contain its own hash, so the two necessarily differ: the content seal is shown for completeness rather than as something reproducible from the file alone.

Hash verified at settlement33 races settledZero outcome fields at freeze

Hypothesis	Races	Wins	Expected	Endpoint
Races with one clear top-rated runner	33	6	8.01	−2.01
Top-rated runner was also the market favourite	17	—	—	−2.31
Top-rated runner was not the market favourite	16	—	—	+0.29

A dash means the settlement record does not carry that field, not that the value is zero. The market-favourite row and the non-favourite row stored the endpoint only.

Prospective record

ACCRUAL

Nothing here is a result yet, and saying so is the point. The sample needed to detect an effect worth detecting is 2,616 races. Until then these are entries in a register, not evidence of anything.

33 / 2,616

Eligible races observed

1.3% of the pre-registered target

Days settled1

Targets issued0

Capital staked£0.00

What has been ruled out

NULL REGISTER

Most of this project's output is negative results, published here because nobody else in the sport publishes them. Each line is a question that was asked properly and answered no.

Does the morning rating add anything once the market price is held constant? adds nothing z = −0.42. 2,487 races with a unique top-rated horse. Residual −0.35pp against race-normalised BSP. Adding the rating to a market-only model made out-of-sample log-loss worse.

Is a horse the market "accepts" a more reliable anchor because our model also picked it? no effect Runners the model accepts win 41.8% of the time; those it rejects, 11.0%. A striking gradient, until you compare it with the market alone, which predicts 40.7% and 12.0%. The same gradient appears for runners the model never touched. The gradient is the price.

Does the 8am rating anticipate where the price moves during the day? no anticipation z = −0.03. Tested against a live positive control: eventual winners shorten at z = +5.57, so the test can see real information. It would have caught a 3.85% average price move. The rating moves it by 0.05%.

Do horses whose price collapses pre-race underperform their final price? no underperformance z = −1.15. Raw steam-fade looked real at −0.91pp. It vanished once compared within the same price band. It was band composition, not overreaction.

Is there value in the place market where the win market is efficient? refused Measured directly: residuals −0.74 to +2.80pp, all inside the confidence interval. Harville conversion permanently refused: it overstates favourites.

Does the draw carry information on its own? killed Powered held-out null at n = 477. Closed, and not reopened.

Three findings that looked significant were killed by the project's own checks before they were reported: a power-recalibration record, a monotone field-size record, and a market-rank cell that failed period replication (−10.01pp in test, +0.03pp in training). They are logged rather than deleted.

The reference market

WHY THIS IS HARD

Betfair's starting price is the benchmark every result is measured against. Across 2,772 races and 25,119 runners carrying a full pre-off price ladder, it calibrates at a slope of 0.9778, statistically indistinguishable from perfect. An independent archive of 1,997 races reproduces that at 0.9737, with a maximum per-band error of 0.96 percentage points. It is a professionalised counterparty, and beating it is the whole difficulty.

To detect a miscalibration of…	Races required
30% relative	181
10% relative	1,625
5% relative	6,499

At α = 0.05 and power 0.80, against a market favourite rate of p ≈ 0.326. Sample size is the bottleneck, not ideas.

Follow the record as it accrues: join the early-access list →

Standing terms

GOVERNANCE

These are enforced in code, not by intention. The alert layer machine-blocks betting language; the freeze refuses to run if a single outcome field is populated on the morning card; a missed market checkpoint is recorded as a gap and never backfilled from a later price.

Capital £0.00 No staking No validated edge No tips sold Retrospective results are hypothesis-generating only No claim of profitability is made anywhere on this page