I have mucked around in the garbage dump of handicapping (the super long shots) long enough. During 111 days of handicapping I separated 7554 horses with a morning line of 15 or more into two groups, those who show some promise (maybe) and those who do not show promise (pass). This should be enough data to satisfy most handicappers, except statisticians.
My original premise was "maybe" horses are superior to "pass" horses and the study supported this contention.
Maybe horses = 1505 picked and 100 won (6.6% hit rate). Loss was $480 for $3010 bet (16.0%). 356 hit the board (23.7% in the money).
Pass horses = 6049 picked and 99 won (1.6% hit rate). Loss was $5317 for $12,098 bet (43.9%). 555 hit board (9.2% in the money).
The maybe horses were only slightly better than the house take while the pass horses were much worse. Maybe horses were no secret to bettors with an average payoff per hit of $25.30, but the pass horses were a mystery, payout per hit of $68.50.
The reason I chose to make these picks public was to keep myself honest. It is so easy to pretend you really would have bet that $50 horse. A public record prevents such nonsense.
__________________
“Man is a slow, sloppy, and brilliant thinker; computers are fast, accurate, and stupid.”
― John Pfeiffer
|