Live Derby Odds

[Dunbar]

Originally Posted by moses

With Serengeti Empress winning, can we fully rely on the double will pays to predict the derby odds?

It's true that there's more room for noise with a long shot having won the Oaks, but there's a lot of data in the exacta pool, and it's pretty convincing that Game Winner will not be in the top 2 in the win pool. It looks close between GW and Max Security for the 3rd fav position.

If you compare exacta payoffs of the 4 of them, here's how often each of them has the lowest payoff when in the 1st position of the exacta: (indicating, most favored)

Imp: 9
Tac: 4
GW: 4
MS: 2

And here's how often each of them had the highest payoff, indicating least favored:

Tac: 0
Imp: 2
MS: 8
GW: 9

Having looked at it more closely, it does look like GW will edge out MS for 3rd favorite.

[moses]

Originally Posted by Dunbar

It's true that there's more room for noise with a long shot having won the Oaks, but there's a lot of data in the exacta pool, and it's pretty convincing that Game Winner will not be in the top 2 in the win pool. It looks close between GW and Max Security for the 3rd fav position.

If you compare exacta payoffs of the 4 of them, here's how often each of them has the lowest payoff when in the 1st position of the exacta: (indicating, most favored)

Imp: 9
Tac: 4
GW: 4
MS: 2

And here's how often each of them had the highest payoff, indicating least favored:

Tac: 0
Imp: 2
MS: 8
GW: 9

Having looked at it more closely, it does look like GW will edge out MS for 3rd favorite.

Good stuff. That all makes sense.

[Dunbar]

Originally Posted by moses

Good stuff. That all makes sense.

Thanks.

You can use a horse's exacta odds to construct the equivalent of a win bet. The odds you get that way will usually be lower than that horse's actual win odds because of the bigger takeout in the exact pool. But doing it that way, I get these odds for those 4 horses:

Imp: 3.9-1
Tac: 4.5-1
GW: 4.6-1
MS: 5.1-1

So, looking at it that way, GW looks closer to Tacitus than I thought.