Stiff sprint stayers on testing-ground tracks: more stamina demand should help more

Reasoning

Why this hypothesis exists: what the lab saw, the question it asked, the mechanism it proposed, what it predicted, and the result that would prove it wrong. Written before the test was run.

1. Observed
   • H181-proven-over-further-at-a-stiff-track (H181: proven over further at a stiff track) was GREEN in the earlier research: A/E 1.08 on 9279 bets (periods 1.06 / 1.07 / 1.12); now PROMISING. O-7cdd45d9
2. AskedIs the sprint edge stronger where the track also tends to ride testing, adding to the stamina demand? answered
3. MechanismA stiff finish on ground that rides testing compounds the stamina demand, so a sprint plays even more like a longer race.
4. Why the market misses itSprint betting keys on early speed and last-time speed figures; the market does not credit stamina when the trip is short.
5. Predictedpositive, A/E 1.20 or more on 120+ bets
6. Falsified ifA/E below the parent sprint system's full-base figure on comparable bets
7. Rule sketchrace_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 1 AND last_win_distance_yards >= distance_yards + 330
8. question idQ-2904981c

Mechanism and rationale

A stiff finish on ground that rides testing compounds the stamina demand, so a sprint plays even more like a longer race.

The rule

race_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 1 AND last_win_distance_yards >= distance_yards + 330

Reads: race_type, distance_yards, track_stiff_finish, track_testing_ground, last_win_distance_yards

Pre-registration

Spec hash (sha256 of the canonical pre-spec)

06eeec491d3395b5f34164750e61716e52eb76c612d4319374ff42bd99b807eb

Frozen

not frozen

The definition can still change.

Confounders

• Few courses are both stiff and testing-ground flagged, so one or two courses (and their draw or pace biases) could carry the result; check by course
• Season: testing ground comes in spring and autumn, and with post-October-2025 data covering one spring a period split is partly a season and surface split
• Price band: the parent family's BSP ROI far above SP ROI points to longshots; judge A/E at 14/1 or shorter
• Today's going: the track flag is a tendency, so any lift may come only on days the ground is actually soft or heavy (refinement 2) and not from the track flag itself
• Class and field size: stiff galloping tracks often stage bigger, better sprint handicaps where trip-droppers are a different kind of horse

Control

{
  "description": "Two comparisons. (1) Same-race rivals: the other runners in these races should score about A/E 1.0 or below; if they also beat their prices, the effect belongs to the courses or race conditions, not to stayers. (2) The question is whether testing-ground tracks add to the parent stiff-sprint edge, so the key comparison is the parent sprint system on stiff tracks that are NOT flagged as riding testing. The hypothesis needs the test rule to beat this control rule's A/E; if it does not, the 'more stamina demand helps more' claim fails even if the test rule beats its prices.",
  "rule": "race_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 0 AND last_win_distance_yards >= distance_yards + 330"
}

Discovery result

fail broad rule · 16 Oct, 12:43

Bets

0

0 winners

A/E

–

winners vs the prices

SP return

–

level stakes

Betfair SP return

–

0 bets with a BSP

Zone: discovery. Evaluator eval-1.0, data 6a4e98be18e83933

Falsification

No attacks yet. Only promising ideas are attacked.

Sealed holdout (vault)

Not queried. Only a frozen idea can spend its one query against the 2020-25 holdout.

Forward record

No forward record. Shadow tracking starts after the vault and counts only races run after the freeze.

Lineage

Agent runs

• designer v1ok · 36s · 16 Oct, 12:42

  Output (schema-checked)

  {
    "rule": "race_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 1 AND last_win_distance_yards >= distance_yards + 330",
    "refinements": [
      {
        "label": "14/1 or shorter",
        "rule": "race_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 1 AND last_win_distance_yards >= distance_yards + 330 AND sp_decimal <= 15"
      },
      {
        "label": "ground actually testing today",
        "rule": "race_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 1 AND last_win_distance_yards >= distance_yards + 330 AND going_group IN ('soft', 'heavy')"
      }
    ],
    "control": {
      "description": "Two comparisons. (1) Same-race rivals: the other runners in these races should score about A/E 1.0 or below; if they also beat their prices, the effect belongs to the courses or race conditions, not to stayers. (2) The question is whether testing-ground tracks add to the parent stiff-sprint edge, so the key comparison is the parent sprint system on stiff tracks that are NOT flagged as riding testing. The hypothesis needs the test rule to beat this control rule's A/E; if it does not, the 'more stamina demand helps more' claim fails even if the test rule beats its prices.",
      "rule": "race_type = 'Flat' AND distance_yards <= 1340 AND track_stiff_finish = 1 AND track_testing_ground = 0 AND last_win_distance_yards >= distance_yards + 330"
    },
    "pass_criteria": {
      "min_bets": 150,
      "min_ae": 1.1,
      "max_p": 0.05
    },
    "forward_bar": {
      "min_bets": 100,
      "min_ae": 1.05,
      "max_p": 0.2
    },
    "confounders": [
      "Few courses are both stiff and testing-ground flagged, so one or two courses (and their draw or pace biases) could carry the result; check by course",
      "Season: testing ground comes in spring and autumn, and with post-October-2025 data covering one spring a period split is partly a season and surface split",
      "Price band: the parent family's BSP ROI far above SP ROI points to longshots; judge A/E at 14/1 or shorter",
      "Today's going: the track flag is a tendency, so any lift may come only on days the ground is actually soft or heavy (refinement 2) and not from the track flag itself",
      "Class and field size: stiff galloping tracks often stage bigger, better sprint handicaps where trip-droppers are a different kind of horse"
    ],
    "mechanism": "A stiff uphill finish makes a 5-6.5f sprint test stamina in the last furlong. On a track that tends to ride testing, the ground adds to that demand, so the race plays even more like a longer trip. A horse whose last win came at least 330 yards further has shown that stamina, and should gain more here than at a stiff track on faster ground.",
    "market_miss": "Sprint betting is driven by early speed, draw and last-time speed figures. A horse whose last win came over further looks too slow for a sprint on those figures, and the market does not adjust for how much a stiff, testing course moves the race toward stamina.",
    "prediction": {
      "direction": "positive",
      "size": "A/E about 1.20 on 150+ bets, above the parent stiff-sprint system on non-testing stiff tracks",
      "falsified_if": "A/E below 1.05 on the broad rule, or A/E no higher than the control rule (stiff sprint stayers with track_testing_ground = 0), or a lift found only at prices above 14/1 or at a single course"
    },
    "notes": "Rule kept as sketched. It uses last_win_distance_yards to match the live parent system (SYS-stayer-dropped-to-a-stiff-sprint), so it is a strict subset of that system. last_win_yards also exists, and whether the two columns match is an open method finding. Both columns are available from 2020-09. The predicted 120 bets is below the lab floor, so min_bets is set at the floor of 150. A thin result that points the right way goes on to falsification as usual. min_ae is raised to 1.10 because the claim is that this subset beats a parent that already shows an edge. Only two refinements are pre-registered, both taken from the reasoning: the price cap, and actual going to separate today's ground from the track tendency."
  }

Timeline

1. Job 1423 finished

   16 Oct, 12:43 · engine:compute-1

2. H-9eff409a Discovery → KILLED fails: 0 bets, needs 150; A/E None, needs 1.1; p None, needs 0.05 or lower; periods [], each needs 1.0 (not a niche pass: only 0 bets; A/E None below 1.1; p None above 0.5)

   16 Oct, 12:43 · engine:evaluator

3. discovery result for H-9eff409a (broad rule): fail

   16 Oct, 12:43 · engine:evaluator

4. H-9eff409a Designed → RESEARCHdiscovery measuring on the discovery zone

   16 Oct, 12:43 · engine:orchestrator

5. compute-1 took job 1423 (evaluate, attempt 1)

   16 Oct, 12:43 · compute-1

6. Job 1420 finished

   16 Oct, 12:43 · engine:ai-1

7. Queued evaluate on the compute lane: discovery run of the broad rule

   16 Oct, 12:43 · engine:scheduler

8. H-9eff409a Proposed → RESEARCHdesigned pre-spec written and hashed

   16 Oct, 12:43 · agent:designer@v1

9. ai-1 took job 1420 (design, attempt 1)

   16 Oct, 12:42 · ai-1

10. Queued design on the ai lane: write the pre-spec

    16 Oct, 12:41 · engine:scheduler

11. New idea H-9eff409a: Stiff sprint stayers on testing-ground tracks: more stamina demand should help more (sharpen)

    16 Oct, 12:41 · agent:investigator@v1

Every change is an event with an actor: an agent and its version, an engine, a human, or the importer. Current stage: Killed.