| |
Racing Forum
no it isn't
Posted By: Dunbar
Date: 16 May 02, 12:39 pm
Doesn't matter if you are trying to win two consecutive races, or trying to win a combination bet in a single race. Your success is directly related to the probabilities of each runner.
It's not the same. In 2 consecutive races, both your win bets CAN win. Therefore it's reasonable to multiply win probabilities.
In a single race, both horses CANNOT win. Therefore it's unreasonable to multiply win probabilities.
And the "fair odds" x "fair odds" x 2 is a very good approximation of your acceptable returns. Optimal... no. Acceptable... yes.
I'm not concerned with optimal or not. I'm concerned with bets that are outright unacceptable, based on your own estimate of fair win odds. Any bet one would make that has negative ev based on one's own assesment of the win odds is unacceptable, IMO.
Now maybe there is some rule of thumb that people use that works reasonably well, but if so, you are not expressing it well. You gave an example of an exacta with 2 horses of fair win odds of 2-1 and 4-1, and you concluded that getting exacta odds of 4-1 (and a $10 payoff) was reasonable. In fact, a fair payoff is $20 or $24, depending on which horse you have put first in the exacta. No one can convince me that $10 is an acceptable return on an exacta that needs at least $20 to be fair. HOWEVER, looking at your "fair odds x fair odds x 2" rule of thumb, maybe you should have calculated 2 x 4 x 2 = $16 as an estimate of the fair payoff. That would have been closer to the $20 value I get using Harville. It's still too low to be acceptable, IMO, but at least it's in the ballpark. Now that I look back, I see that is what you were doing in your original post with your Wednesday picks. But you made a mistake (in your own method) when you gave the 2-1, 4-1 example.
As usual when there is a good discussion, I've learned a good bit.
--Dunbar
- no it isn't -- Dunbar -- 16 May 02, 12:39 pm
| |
Racing Forum is maintained by Pi Yee Press