H76: course likes class drop

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

   H76: course likes class drop KILLED red

   because Down in class at a course where such runners have beaten the market by 25%+ before. Lower-grade tracks where a class dropper really does outclass them.

   cp_class_drop_runs >= 60 AND cp_class_drop_wins >= 1.25 * cp_class_drop_exp

   made by import

2. Tested15 Oct, 23:10

   broad: A/E 1.05 on 1,346 bets, by period 1.05 · 1.00. info

   Unstable: periods range 1.00 to 1.05. Zone: all. On H76: course likes class drop

   By period

   betsA/E0.71.01.3

   2020-221,3001.05

   2023 to a year ago461.00

   Last year0–

3. Tested15 Oct, 23:10

   100+ earlier runners: A/E 1.02 on 1,007 bets, by period 1.02 · 1.00. info

   Unstable: periods range 1.00 to 1.02. Zone: all. On H76: course likes class drop

   By period

   betsA/E0.71.01.3

   2020-229611.02

   2023 to a year ago461.00

   Last year0–

4. Tested15 Oct, 23:10

   jumps: A/E 1.04 on 93 bets, by period 1.08 · 1.00. info

   Unstable: periods range 1.00 to 1.08. Zone: all. On H76: course likes class drop

   By period

   betsA/E0.71.01.3

   2020-22471.08

   2023 to a year ago461.00

   Last year0–

5. Tested15 Oct, 23:10

   Flat: A/E 1.05 on 1,253 bets, by period 1.05. info

   Every period above 1.00 (spread 0.00). Zone: all. On H76: course likes class drop

   By period

   betsA/E0.71.01.3

   2020-221,2531.05

   2023 to a year ago0–

   Last year0–

6. Killed15 Oct, 23:10

   H76: course likes class drop Discovery → KILLEDimport: tier C: rejected, do not test again unless new data changes the idea; best [broad] RED A/E 1.047 on 1346 bets (broad RED)