H10: yard good at hurdle to chase

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

   H10: yard good at hurdle to chase KILLED amber

   because Hurdle run then a chase, from a yard whose earlier runners making that switch beat the market by 30%+.

   tx_h2c_runs >= 15 AND tx_h2c_wins >= 1.3 * tx_h2c_exp

   made by import

2. Tested15 Oct, 23:10

   broad: A/E 1.05 on 1,712 bets, by period 1.15 · 0.97 · 1.10. info

   Betfair SP return +43.6%. Unstable: periods range 0.97 to 1.15. Zone: all. On H10: yard good at hurdle to chase

   By period

   betsA/E0.71.01.3

   2020-224861.15

   2023 to a year ago9000.97

   Last year3261.10

3. Tested15 Oct, 23:10

   has chased before: A/E 1.08 on 590 bets, by period 1.30 · 0.85 · 1.45. info

   Betfair SP return +60.4%. Unstable: periods range 0.85 to 1.45. Zone: all. On H10: yard good at hurdle to chase

   By period

   betsA/E0.71.01.3

   2020-221271.30

   2023 to a year ago3240.85

   Last year1391.45

4. Tested15 Oct, 23:10

   handicap: A/E 1.04 on 1,291 bets, by period 1.14 · 0.96 · 1.11. info

   Betfair SP return +50.0%. Unstable: periods range 0.96 to 1.14. Zone: all. On H10: yard good at hurdle to chase

   By period

   betsA/E0.71.01.3

   2020-223671.14

   2023 to a year ago6730.96

   Last year2511.11

5. Killed15 Oct, 23:10

   H10: yard good at hurdle to chase Discovery → KILLEDimport: tier C: rejected, do not test again unless new data changes the idea; best [has chased before] AMBER A/E 1.078 on 590 bets (broad AMBER)