IQ54: failed to finish jumping as usual

The notebook

In order: what the lab saw, what it asked, what it proposed and why, what the tests said, and what an investigator concluded. Numbers are aggregates on the discovery data unless a step says otherwise.

1. Hypothesised15 Oct, 23:10

   IQ54: failed to finish jumping as usual SHADOW

   because Failed to finish last time, in today's code, with a Jump Index no more than 0.15 below its own earlier average. Nothing unusual in the jumping: it was looked after, and the run can be ignored.

   race_type IN ('Chase', 'Hurdle') AND last_position IS NULL AND l_same_code = 1 AND l_jump_index - s_jump_index_prev >= -0.15

   made by import

2. Tested15 Oct, 23:10

   HOLDOUT 2023-25, all: A/E 1.13 on 2,028 bets, by period 1.13. info

   Every period above 1.00 (spread 0.00). Zone: all. On IQ54: failed to finish jumping as usual

   By period

   betsA/E0.71.01.3

   2020-220–

   2023 to a year ago2,0281.13

   Last year0–

3. Tested15 Oct, 23:10

   Britain: A/E 1.13 on 1,599 bets, by period 1.13. info

   Every period above 1.00 (spread 0.00). Zone: all. On IQ54: failed to finish jumping as usual

   By period

   betsA/E0.71.01.3

   2020-220–

   2023 to a year ago1,5991.13

   Last year0–

4. Tested15 Oct, 23:10

   Ireland: A/E 1.14 on 429 bets, by period 1.14. info

   Every period above 1.00 (spread 0.00). Zone: all. On IQ54: failed to finish jumping as usual

   By period

   betsA/E0.71.01.3

   2020-220–

   2023 to a year ago4291.14

   Last year0–

5. Tested15 Oct, 23:10

   discovery year (reference): A/E 1.11 on 1,261 bets, by period 1.11. info

   Betfair SP return +10.6%. Every period above 1.00 (spread 0.00). Zone: all. On IQ54: failed to finish jumping as usual

   By period

   betsA/E0.71.01.3

   2020-220–

   2023 to a year ago0–

   Last year1,2611.11

6. Promoted15 Oct, 23:10

   IQ54: failed to finish jumping as usual Falsification → FROZENimport: frozen

7. Promoted15 Oct, 23:10

   IQ54: failed to finish jumping as usual Vault → SHADOWimport: holdout SUPPORTIVE: A/E 1.129 on 2028 (GB 1.128, IRE 1.136), p 0.0737 against BH 0.003448 over 29 tests; LEARNINGS read it as holding

8. Tested15 Oct, 23:21

   forward vs bar {'min_bets': 100, 'min_ae': 1.0, 'max_p': 0.2} on 0 bets. info

   Zone: forward. On IQ54: failed to finish jumping as usual

9. Observed16 Oct, 10:10

   IQ54-failed-to-finish-jumping-as-usual (IQ54: failed to finish jumping as usual) pointed the predicted way (positive) on the sealed holdout but too weakly to prove it: A/E 1.129 on 2028 bets (p 0.074, false-discovery threshold 0.0034); tracked forward in SHADOW.

   result0.35waiting for a reasoning passfrom IQ54-failed-to-finish-jumping-as-usual

10. Tested16 Oct, 23:21

    forward vs bar {'min_bets': 100, 'min_ae': 1.0, 'max_p': 0.2} on 0 bets. info

    Zone: forward. On IQ54: failed to finish jumping as usual