Keeneland lowers DD takeout to 15% from 22%

[lamboguy]

when it comes to takeout and breakage (kentucky not included) one has to know how much money is needed in the course of a day to make a handle. just for argument’s sake for this example i am going to use the total handle for the day to be $18 million and the amount of takeout and breakage to remain constant which for nothing easier to use constant numbers and i am going to use round numbers and 20% total takeout.
walking into the windows there is $4 million total that can possible be lost on a given race card.

we start race by race

Race 1 $4,000,000 remaining after race $3,200,000

Race 2 $3,200,000 remaining after race $2,600,000

Race 3 $2,600,000 remaining after race $2,100,000

Race 4 $2,100,000 remaining after race $1,700,000

Race 5 $1,700,000 remaining after race $1,400,000

race 6 $ 1,400,000 remaining after race $1,100,000

race 7 $1,100,000 remaining after race $800,000

race 8 $800,000 remaining after race $600,000

race 9 $600,000 remaining after race $400,000



Bottom line using very rough numbers, $4 million walks in the door, and after 9 races $400,000 comes home after a day at the races.

[Redboard]

Keenland Saturday 4/6. $1 bets.

Race DD EX
1 N/A $39.42
2 $28.92 $22.25
3 $13.79 $11.34
4 $11.48 $33.55
5 $35.49 $32.37
6 $32.72 $24.83
7 $55.17 $441.97
8 $45.98 $9.60
9 $22.38 $18.40
10 $16.17 $9.96
11 $7.45 $14.90
Average $26.96 $59.87

On Saturday, the average payout of the exacta was over twice that of the DD.

As I mentioned earlier, the DD payout never impressed me, which is wht I don't play it. Now you might say that Race 7 was an outlier with that huge Exacta payout which skewed the analysis. But, there's a better chance that a longshot gets second place than first.

If Kee slashes the Ex down to 15%, wake me up.

[AndyC]

Originally Posted by Redboard

Keenland Saturday 4/6. $1 bets.

Race DD EX
1 N/A $39.42
2 $28.92 $22.25
3 $13.79 $11.34
4 $11.48 $33.55
5 $35.49 $32.37
6 $32.72 $24.83
7 $55.17 $441.97
8 $45.98 $9.60
9 $22.38 $18.40
10 $16.17 $9.96
11 $7.45 $14.90
Average $26.96 $59.87

On Saturday, the average payout of the exacta was over twice that of the DD.

As I mentioned earlier, the DD payout never impressed me, which is wht I don't play it. Now you might say that Race 7 was an outlier with that huge Exacta payout which skewed the analysis. But, there's a better chance that a longshot gets second place than first.

If Kee slashes the Ex down to 15%, wake me up.

What is the significance of a higher average payout? Don't the exactas have significantly lower probabilities of hitting?

[lamboguy]

Originally Posted by AndyC

What is the significance of a higher average payout? Don't the exactas have significantly lower probabilities of hitting?

i don't know about payoffs, but the man is right about lower takeout in exacta's.

[thaskalos]

Originally Posted by Redboard

Keenland Saturday 4/6. $1 bets.

Race DD EX
1 N/A $39.42
2 $28.92 $22.25
3 $13.79 $11.34
4 $11.48 $33.55
5 $35.49 $32.37
6 $32.72 $24.83
7 $55.17 $441.97
8 $45.98 $9.60
9 $22.38 $18.40
10 $16.17 $9.96
11 $7.45 $14.90
Average $26.96 $59.87

On Saturday, the average payout of the exacta was over twice that of the DD.

As I mentioned earlier, the DD payout never impressed me, which is wht I don't play it. Now you might say that Race 7 was an outlier with that huge Exacta payout which skewed the analysis. But, there's a better chance that a longshot gets second place than first.

If Kee slashes the Ex down to 15%, wake me up.

The comparison was obviously skewed by that huge exacta payoff. If you look at the races individually, out of the 10 races with available wagering data, the DD payoffs were larger in 7 of them.

[Poindexter]

A double is a parlay from the winner of one race to the winner of the next race basically. An exacta is a parlay from the winner of the race to the second place finisher in the same race. Which as a rule (obviously not as accurate as it once was due to the Caw) typically pays Win price * place price of 2nd place finisher. So if exactas are going to pay better than doubles that means that the place price of the second horse will exceed the win price of the next race winner and of course vice versa. So Anybody with a database can look at average win prices and average place prices of 2nd place finishers from a track to see what will pay better typically. As Andy C basically is saying it all relative to their chances of coming in.

But obviously players have their strengths and preferences in pools. So when tracks cherry pick which pools to reduce takeout in, they are depriving some bettors to fully capitalize on the benefit. So I completely understand the sentiment of Redboard and Lambo. Basically Keenland is saying that if you play their doubles you get a 9% bonus on your payout. If you play exactas you don't. If you are much better in exactas than doubles, that doesn't seem fair to you, because it isn't. By the way Keenland has a 19.5% takeout on exactas. So by bringing them down to 15% would have only provided a bonus of about 5.5%. If this works out for them, I would like to see them bring both down next time.

[iamt]

For the last 10 years the average double payoff is slightly higher than the exacta at Keeneland, less than 2%. Across all tracks the difference is greater ~15%. Keeneland being closer may just be a product of the different takeouts as there should be more betting combinations on average in a double than an exacta.

[MJC922]

IMO players should probably be incentivized to wager in the pool where they have the best chances for longterm success. This doesn't even have to be guessed at, it could be studied. My personal records in recent years show the DD is the correct pool for doing that. Someone elses records may show the exacta, however I think it's pretty safe to say that if a novice can't make money in doubles it isn't going to get better in any other pool. Even for the purist like me in the win pool with projected odds available (the latter KEE hasn't made available yet have they?) IMO has a bit higher money management bar for the novice.

Increased churn should equal increased handle and fewer people leaving flat broke should equal increased handle on tomorrow's card. While it may make sense to say low takeout should be applied on every type of wager it may in practice actually lead to a better bottom line for both the tracks and the players when 'guard rails' are put up by way of using takeout to funnel wagers into specific types of bets like this.

In theory I can see where there might be a spreadsheet that could be used to optimize and balance all of the desired outcomes. Such an optimization might end up showing different takeouts should be implemented in the different pools, not necessarily one low takeout applied across all pools. I suspect some of you might in fact not like to see what types of takeouts something like that would spit out for longer horizontals and all of these blind pools, it wouldn't surprise me if it spit out the notion that there should be a cigarette tax placed on that stuff. It probably wouldn't be a deal breaker anyway because the carryover can eventually neutralize takeout at least for the pick six.

[Redboard]

Originally Posted by thaskalos

The comparison was obviously skewed by that huge exacta payoff. If you look at the races individually, out of the 10 races with available wagering data, the DD payoffs were larger in 7 of them.

But isn't that number the payoff we are playing for? (the skewed one) It doesn't matter whether you lost 99 out of 100 bets. The bottom line is, at the end of the day, it's what you have left in your account.

Now on further analysis, over the long haul, the DD might be the more profitable bet. As I mentioned, it's never been my cup of tea. I've just never won much from it even when I hit it. As Point said, it's about strengths and preferences.

[classhandicapper]

In theory the double should pay more.

The double is win to win.

The exacta can be thought of similar as win to win "if the real winner wasn't in the race". So it's also a kind of parlay but with one less horse in the 2nd race making the average payoff slightly lower.

[thaskalos]

Originally Posted by classhandicapper

In theory the double should pay more.

The double is win to win.

The exacta can be thought of similar as win to win "if the real winner wasn't in the race". So it's also a kind of parlay but with one less horse in the 2nd race making the average payoff slightly lower.

But as a counterpoint, it's more difficult to find the place horse in race 1 than it is to find the winner of race 2. At least it is for me. And it's even more difficult to find the show horse in race 1. I can pretty much spot the horses with winning potential pretty quickly. But the further down I go in the vertical wager...the more complicated the handicapping task becomes.

[classhandicapper]

Originally Posted by thaskalos

But as a counterpoint, it's more difficult to find the place horse in race 1 than it is to find the winner of race 2. At least it is for me. And it's even more difficult to find the show horse in race 1. I can pretty much spot the horses with winning potential pretty quickly. But the further down I go in the vertical wager...the more complicated the handicapping task becomes.

I agree.

I tend to favor exactas because I have so few opinions where I feel confident I am getting good value, there are even fewer opportunities for me to hook value to value horizontally to leverage my opinions. I'd rather try to crush one race vertically. When I hate a short priced horse in single race I can key against him both to win and in the exacta (and tri/super if appropriate). One of my more common bets when I hate a short priced horse and have the race narrowed down to 2 contenders is betting both to win, boxing them, and occasionally using someone else underneath both. There's often a lot of value in getting the favorite out of the exacta.

[AndyC]

Originally Posted by classhandicapper

In theory the double should pay more.

The double is win to win.

The exacta can be thought of similar as win to win "if the real winner wasn't in the race". So it's also a kind of parlay but with one less horse in the 2nd race making the average payoff slightly lower.

That's assuming that the second leg of the double has at least the same number horses as the first leg.

[classhandicapper]

Originally Posted by AndyC

That's assuming that the second leg of the double has at least the same number horses as the first leg.

This is how I am visualizing it.

Imagine 2 fields of 8.

The double is 1 out of 8 parlayed to 1 out of 8.

The exacta is kind of 1/8 (top horse) parlayed to 1 out of 7 (for 2nd) where the winner's probability/pool money is subtracted to get the probability/price of the 2nd horse.

[Poindexter]

If the average field size is 8 than the average number of double combos will be around 64. The average number of exacta combinations will be closer to 56 (over 12% less). The reason the average payoffs are much closer has to do with the fact that longshots in general have a much better chance of coming 2nd than 1st. So if someone with a database tells you how many horses win at over 10-1 and how many horses over 10-1 come 2nd, the latter will be a significantly higher number. So that is why there is not that big a difference between average double payoff and average exacta payoff despite the fact that you have more combos in the double pools.