Dutching Win Bets

[formula_2002]

Originally Posted by ubercapper

I'll try to not be too long winded here. On one of my blogs Saturday, for the Ginger Brew Stakes (Race 9) I recommended a win bet on either/both She's My Type and Sunset Promise at 3/1 or more, adding a place bet at 6/1 or more.

She's My Type went off at 9 to 1 and Sunset Promise at 7 to 1 so both were acceptable bets.

My intent, as it usually is, was to help mostly casual (read "weekend") bettors make a profit. I know dutching to win guarantees one loser even if right.

Ignoring the place bet recommendation for now and just discussing the win bet recommendation, I've been lectured win bet dutching is sub-optimal and as someone who either is an "expert" or "paid" I should know better. There were other, nastier, words used as well.

In any event, ignoring the intent of my blogs being geared to a litany of types of racing fans who I believe are happy to trade some ROI for volatility, I reached out to an old friend who has written on the topic.

Here's what he wrote: "One advantage of dutching is you increase the number of cashes, which cuts volatility. As the odds of each horse increases, your chances of losing streaks increase as well. While in the ideal world we would make perfect odds lines and would make bets after all the money is in the pools, it will never happen. So if you have two horses at 3-1 on your line (assuming it's a perfect line) and one is 7-1 and the other is 9-1, of course you would have a greater ROI if you just bet the 9-1 shot, except that your line won't be perfect and the 9-1 shot with one minute to go might be bet down to 6-1 while the 7-1 horse might drift up to 11-1, which messes up your strategy. Since you can never bet with the final odds, yes, I would dutch both horses to win, with SLIGHTLY more on the horse who's 9-1 as late as possible."
"An overlay is still an overlay, regardless of whether there's another overlay in the race."

The last sentence is the biggest takeaway for me, and I don't understand why some people don't get it and insist that their method of betting just one horse to win is the best method for everyone.


Knowing (hoping?), this board is more kind than Twitter these days, I wanted to hear what some of you have to say on this topic.

Betting more than one horse in a races should mean that each is an overlay.
The correct bet size is in proportion to the odds and the overlay factor.

If I was a betting man and had an edge , and knew the final odds, I would then normalize all plays such the total bet always equals 1.

a dutch play returns the same amount regardless which horse wins.. If I recall correctly, regardless which way you bet it, the $1 returns are the same.

the real fun begins with exacta play. I calculate the "True" exacta probability and the probability the public assigns. Then I compare the two and post the combinations that are overlaid. It's even more fun when you start to create formulas that adjust those bet sizes (probability)..like I say if I was a betting man and had an edge.

[davew]

for betting your 2 horses 3/1 or better suggest you felt each had more than a 25% chance of winning.

if you bet them both for $2, you will cash 50% of the time (if your numbers are decent)

is betting $4 to win $16+ or $20+ over half the time not a killer bet?


whether your 25% win probabilities need adjusted on 'dead in the pool horses' is another debate.

[thaskalos]

Originally Posted by davew

for betting your 2 horses 3/1 or better suggest you felt each had more than a 25% chance of winning.

if you bet them both for $2, you will cash 50% of the time (if your numbers are decent)

is betting $4 to win $16+ or $20+ over half the time not a killer bet?


whether your 25% win probabilities need adjusted on 'dead in the pool horses' is another debate.

If you bet two horses at odds of 3/1, and one of them wins...you have invested $4 to win $8+...not "$16+ or $20+".

[davew]

Originally Posted by thaskalos

If you bet two horses at odds of 3/1, and one of them wins...you have invested $4 to win $8+...not "$16+ or $20+".

the example given at the start of this thread had the horses going off at 7/1 and 9/1 - you stand corrected

[cj]

Originally Posted by davew

for betting your 2 horses 3/1 or better suggest you felt each had more than a 25% chance of winning.

if you bet them both for $2, you will cash 50% of the time (if your numbers are decent)

is betting $4 to win $16+ or $20+ over half the time not a killer bet?


whether your 25% win probabilities need adjusted on 'dead in the pool horses' is another debate.

I'll add if you are willing to take the horses at 3-1, you think they have greater than 25% win probability I would hope, probably at least 35% each. If you only though 25% it is a break even proposition and not worth a play.

Whether there is ever a race where two 3-1 shots actually have a 70% chance of winning is a different question for another day.

[Nitro]

Originally Posted by Michael

Now that I see how you structured your bet... your boat has a few holes in it if you ask me.

Hypothetically: you're saying it's ok to bet both at 3/1. If we run it.... You're saying your 58.5% at 17% take and that's to break even. Highly unlikely this is anywhere near reality.

I don't mean to be rude.. but seems bad.

Using your breakdown I would’ve quit playing a long time ago! I’m not sure which tracks you’re playing, but those I play take their share (17% or whatever) from the entire Win pool before the odds are posted.
So if the tote indicates a 3/1 return, it is in fact a minimum of $8.00 on a $2 wager. The final payout is certainly not $8.00 less whatever the take-out might be!

Originally Posted by thaskalos

If you bet two horses at odds of 3/1, and one of them wins...you have invested $4 to win $8+...not "$16+ or $20+".

Exactly!
And your actual profit of $4 provides you with a decent 100% Profit margin!

Originally Posted by cj

I'll add if you are willing to take the horses at 3-1, you think they have greater than 25% win probability I would hope, probably at least 35% each. If you only though 25% it is a break even proposition and not worth a play.

Whether there is ever a race where two 3-1 shots actually have a 70% chance of winning is a different question for another day.

I’ll submit that “thinking” about a “win probability” or even presuming some “percentage of winning” is purely a subjective scrutiny based on the interpretation of the data that’s being used.

Rather than that, I would suggest relying on a personal betting record of successful Dutch wagering based on an actual hit frequency over a period of time. That frequency should determine objectively how much of a profit margin would be acceptable for future bets and profitability.

In other words if you’re betting on 2 entries at 3/1 with an anticipated return of a 100% profit margin and you’re hitting these Dutch bets say 50% percent of the time, I’d say it’s a very good bet.

On the other hand if your contemplating betting on 2 entries with 8/5 and 5/2 odds the profit margin is only 49% . I would not recommend making that type of bet only because the predetermined and recorded O.A. profit margin is only 50%. Common sense dictates that playing bets with a lower profit margin then the actual hit frequency is a losing proposition over the long haul.
.

[thaskalos]

Originally Posted by davew

the example given at the start of this thread had the horses going off at 7/1 and 9/1 - you stand corrected

That's what the original poster said...but you gave an example of two horses at 3/1...who had a combined fair-odds winning expectation of 50%. If you were talking about the 7/1 and the 9/1 horses that the OP used...where did you get that 50% winning expectation for the wager?

[thaskalos]

Originally Posted by davew

for betting your 2 horses 3/1 or better suggest you felt each had more than a 25% chance of winning.

if you bet them both for $2, you will cash 50% of the time (if your numbers are decent)

is betting $4 to win $16+ or $20+ over half the time not a killer bet?


whether your 25% win probabilities need adjusted on 'dead in the pool horses' is another debate.

See? If the odds are 7/1 and 9/1...how do you figure to win "over half the time"?

[AndyC]

Originally Posted by thaskalos

See? If the odds are 7/1 and 9/1...how do you figure to win "over half the time"?

He's a REALLY good handicapper!

[davew]

Originally Posted by thaskalos

See? If the odds are 7/1 and 9/1...how do you figure to win "over half the time"?

I was using the criteria given in the original post

[lawmaker1]

Assuming that we are all decent handicappers, what is the best book out there on how to bet!

[ubercapper]

Originally Posted by AndyC

He's a REALLY good handicapper!

Andy, I recall our days many years ago on the message boards (AOL?) and hope you're well.

No, I'm not a really good handicapper. Perhaps better than average, and I believe (and the market has agreed) what I do fills a need, but I'll leave it at that.



As for line making, I'm definitely average at best because I never make a line on every horse and use the philosophy Mitchell taught me, which was to make a line to 75% or 80%.

As mentioned in the OP, I thought each horses had a 25% chance to win so fair odds were 3/1 each. I know my line isn't perfect and I also know the odds might shift (perhaps considerably) near post time. However, with odds of 7/1 and 9/1 near post time, even considering my fair odds might have been 4/1 and 9/2 on each, I still would have dutched win bets.

[ubercapper]

Originally Posted by BarchCapper

I enjoy both Ubercapper's AND his Twitter "opponent"s content. There's a really good 2-part interview with the latter in the archives of Jason Beem's podcast - these remain his most-downloaded episodes. If you're interested, they can be found by googling "Inside the Pylons BetAmerica."

Ubercapper, I do appreciate your taking the highest road possible in the full exchange, which I was glad to read . The two of you were pretty much talking past each other. He seemed totally focused on overall ROI, and ignored the volatility aspect, which you were trying to emphasize.

There's a reason good investment professionals hand out those risk tolerance assessments. Some can handle the wild swings of lower W%, higher ROI wagering, without it affecting the way they approach the next race in the middle of a long streak of losers. Many, including myself, would not be able to handicap as well when faced with that.

First, thanks for the kind words.

Second, taking the high road on Twitter is something I strive for, although I'll admit I use the "create a tweet then delete it approach" to avoid sending something which I feel is beneath me. Occasionally I delete a tweet after realizing it would have been better to remain silent. To this day, after nine years on that medium, I have only blocked seven people, although there are about 10 times that which are muted.

Last, I don't understand why the issue of risk tolerance being irrelevant to a lot of bettors, or to put it another way, much less relevant than having fun or making even a small profit, is just so foreign to some people to they point they are downright disparaging to anyone who doesn't feel as they do. The same people don't understand why EV doesn't matter to most of the audience on social media (assuming I'm right the teams and whales don't have Twitter accounts).

[ubercapper]

Originally Posted by lawmaker1

Assuming that we are all decent handicappers, what is the best book out there on how to bet!

I think an oldie but goodie is "Money Secrets at the Racetrack" by Barry Meadow.

[pandy]

I have one customer who told me that he became a consistent winner once he started betting two horses a race, but he tries to to avoid heavy favorites. Basically he breaks each race down to four contenders and bets the two highest odds horses.

I don't knock anything that works for someone. And, as others have said, when you play exacta boxes, you are betting more than one horse in the race.

Personally, I've always felt that an exacta box is a better hedge than betting two horses to win, but I've never done any serious testing on betting two horses to win in the same race. If you bet $20 to win on two horses, I would think that $40 would give a higher ROI over the long run if you keyed those horses in exactas with your other two top contenders... for instance, say the two key horses are the 1 and 2 and your other contenders are the 3 and 4. $6 exacta box 1-2, $7 exacta boxes 1 and 2 with 3 and 4, total bet $40.00. Also, if you get rebates, the higher rebates on exacta bets are another hedge.