2 Horse Win Betting

[coach_lowe]

When is it appropriate?

[cmoore]

When YOU think the favorite is vulnerable. This is a great time to bet 2 horses in a race.

[limewithsoco]

if u wager on 2 horses u are cutting your odds in half.. example: $5 to win on a 10-1 and $5 to win on a 20-1.. if u hit the 20-1 u get back $100 (which is nice) BUT you put up 10 to get 100 (only 10-1 back on your money with a 20-1 shot). if u play 2 horses in a race u are better off in my opinion doing daily doubles or pick 3s.

[JohnGalt1]

Betting 2 horses to win is similar to betting an entry--but it's you who decide who the entry is.

I'll only do it if I like 2 horses equally, and the odds total at least 8-1. Ex. 6-1 & 2-1.

If I think one is slightly better I'll bet 60% on one and 40% on the other. The second horse may be a longshot angle horse like first claim blinkers off 12-1.

An example of an entry I made is CBY race 7 Sat Claiming Crown Glass Shipper I'll bet #1 Thunder Belle @ML 6-1 and #4 Miranda Diane @ML 4-1 both equally to win and play a 145 ex box.

[cj's dad]

Perfect example for betting 2 in the same race:

Yeasterday's race #7 @ Monmouth; I liked both the #2 & #6, M/L odds of 6-1 and 20-1, bet both, they went off at odds of 11-1 and 31-1.

The #6 wins and pays $65.80

I see it no differently than playing a horse to win and then putting that same horse 2nd to others in an exacta. Only one wager can win.

[jonnielu]

When is it appropriate?

Never, in today's game you increase your chance of winning very little, while you concede to lose at least one bet. Also, you would only consider doing this with 2 horses at longer odds, therefore, if you bet both to place, your chances to cash both tickets increase.

The method of win betting multiple horses in today's game is the DD/Pk3/Pk4 wagers.

jdl

[Irish Boy]

Never, in today's game you increase your chance of winning very little, while you concede to lose at least one bet. Also, you would only consider doing this with 2 horses at longer odds, therefore, if you bet both to place, your chances to cash both tickets increase.

Why would you bet two horses to place? That makes almost no sense.

1.) Even with two win bets, hitting one is likely to pay as much or more than both place bets. Two $10 win bets on 10-1 shots returns $100 (plus the $10 you bet). You need a $10 place payoff on both horses to equal that, and you need to hit both. Neither is likely to happen. And every time you hit both place bets, you would have hit a win bet. It certainly isn't true that every time you hit a win bet you would have hit both place bets.

2.) The higher the odds on your win bets, the crappier your place payouts (comparatively speaking). That 40-1 shot you were thinking of including is only going to pay a fraction of that in the place pool, because this is where uninformed tourist money really takes a toll on the small pools.

3.) Rather than bet $20 on two separate horses to place, why not just have a $2 saver exacta box? If you hit those two $10 place bets, you'd be betting $20 to win maybe $100. If the exact same result happens and you bet into the exactas instead, you'll almost certainly win over $100 for only $4 dollars.

4.) The only time this will hurt is when one of your horses finishes second and the other finishes third or worse. However, double place betting is still a bad idea here for all the reasons place betting is normally a bad idea (longshots become awful underlays, breakage erodes returns more, smaller pools, etc.)

[jonnielu]

Why would you bet two horses to place? That makes almost no sense.

1.) Even with two win bets, hitting one is likely to pay as much or more than both place bets. Two $10 win bets on 10-1 shots returns $100 (plus the $10 you bet). You need a $10 place payoff on both horses to equal that, and you need to hit both. Neither is likely to happen. And every time you hit both place bets, you would have hit a win bet. It certainly isn't true that every time you hit a win bet you would have hit both place bets.

2.) The higher the odds on your win bets, the crappier your place payouts (comparatively speaking). That 40-1 shot you were thinking of including is only going to pay a fraction of that in the place pool, because this is where uninformed tourist money really takes a toll on the small pools.

3.) Rather than bet $20 on two separate horses to place, why not just have a $2 saver exacta box? If you hit those two $10 place bets, you'd be betting $20 to win maybe $100. If the exact same result happens and you bet into the exactas instead, you'll almost certainly win over $100 for only $4 dollars.

4.) The only time this will hurt is when one of your horses finishes second and the other finishes third or worse. However, double place betting is still a bad idea here for all the reasons place betting is normally a bad idea (longshots become awful underlays, breakage erodes returns more, smaller pools, etc.)

Thanks, you know, I've never stepped back to really consider how the results of horse races always make crystal clear sense before. Now I see the error of my ways.

jdl

[dutchboy]

My simple minded question of the day is?

What difference does it make if you bet two horses in one race at good odds or bet one horse in two races at good odds.

[DJofSD]

When is it appropriate?
It is appropriate when you can make a profit regardless of which horse wins.

You can make a profit by dutching to the odds. Dutching to the odds means betting an amount that is scaled to the current odds, a larger amount on the shorter odds and a lesser amount on the longer odds one. There are dutching tables, or, software programs that will give you the specifics. I don't know if George Green still offers it but there used to be a program named HedgeHog.

Why bet two horses to win? It certainly is not going to be popular with some bettors. The late Doc Sartin advocated betting multiple win bets. I believe he taught it more as a psychological ploy more than anything else. Going to the window to cash more often than you would only betting one horse can do wonders for your confidence.

If at the end of the day, you're kicking yourself with a lot of would've, should've, could've self talk, consider dutching two horses to win before doing anything more.

[andymays]

I can't say never do it. But almost never do it!

I've done it maybe 3 times in my life.

[Valupix]

If you don’t look at the game in terms of percentages and value you’re probably going to view my reply just as silly as I view the replies geared towards “never bet more than one horse” and “don’t because it reduces your odds.”

Let’s say you have a horse you think has a 40% chances of winning a given race. You’re at post time and the horse is showing at 5/2 odds on the board. Theoretically you only need 3/2 to break even. So the 5/2 is a solid overlay price.

But there is also another horse deeper in the field that you have assessed with a 10% chance of winning. This horse is showing at 15/1 odds on the board. You only need 9/1 to break even. Another solid overlay. Are you telling me you would pass up a bet on the second overlay? No way!

Combined betting the two puts you in a situation to have a 50% chance of winning the race. Outside of late scratches, surface condition changes or some other relevant late change in the race situations your 40% assessment on your top horse, nor the 10% assessment on the other is going to change. You haven’t reduced any chances of winning or losing this particular race. Not even if there were a third, or even fourth or more overlay horse(s) that you also included in the bets. The assessments aren’t changing, but the prices will. It’s just a matter of price exceeding you assessment each horses individual likelihood of winning the race.

Of course that’s based on you ability to make good assessments. But if you make then, you gotta trust them. It’s also my opinion and in the end I may have no clue what the heck I’m talking about. (the old don’t beat me up if you disagree disclaimer)



　

[Valupix]

Collectively your assessments on the entire field is never going to exceed 100% combined. In theory you can bet any and all overlays and never be create situations of disadvantage.

Theory being references to your own ability to make good assessments. If you make bad judgments on horses chances of winning then you’re always going to being at a disadvantage.

[ezrabrooks]

Perfect example for betting 2 in the same race:

Yeasterday's race #7 @ Monmouth; I liked both the #2 & #6, M/L odds of 6-1 and 20-1, bet both, they went off at odds of 11-1 and 31-1.

The #6 wins and pays $65.80

I see it no differently than playing a horse to win and then putting that same horse 2nd to others in an exacta. Only one wager can win.

I played the race...and it would have taken me more than a Win Box to get to the #6...

Ez

[cj's dad]

Sorry you missed it, I didn't.

[Tom Barrister]

When is it appropriate?

It's appropriate if both horses are overlays. It isn't appropriate otherwise.

[ezrabrooks]

Sorry you missed it, I didn't.

CONGRATS...

[lansdale]

When is it appropriate?

coach lowe,

Please ignore any advice in this thread telling you to bet only one horse. Many areas in this game are a matter of opinion, but this one is not. Anyone with even the slightest grasp of the mathematics of betting knows that only betting overlays is the way to go, regardless of how many you bet. Others in this thread have expressed this, but Valupix has stated the reasoning behind it most clearly. If you're looking for a more extensive study of correct wagering, get a copy of Dick Mitchell's 'Commonsense Betting', in which he lays out his technique for betting overlays. Of course, as Valupix says, this assumes that you're capable of making a reasonably correct line for the race - of determining which horses are genuine overlays. In my own handicapping, I would generally say, though that there is more than one real overlay in maybe 25% or 30% of races, but YMMV. Best of luck.

Cheers,

lansdale

[DJofSD]

Sorry you missed it, I didn't.That's cold. So, how much did you win?

[bisket]

i do this if the odds warrant it. most times if i do this i've also got an exacta or trifecta wager on the race also. the best race of all to do this is the derby. the odds in that race are always nowhere near what they should be. you can usually figure a horse who's usually 2-1 in any other race is 4-1...... and so on. it only makes sense to do it when betting that race.

[pktruckdriver]

Thanks, you know, I've never stepped back to really consider how the results of horse races always make crystal clear sense before. Now I see the error of my ways.

jdl


Wow the truth comes out, Yes Sir.


Patrick

[dutchboy]

It is appropriate when you can make a profit regardless of which horse wins.

You can make a profit by dutching to the odds. Dutching to the odds means betting an amount that is scaled to the current odds, a larger amount on the shorter odds and a lesser amount on the longer odds one. There are dutching tables, or, software programs that will give you the specifics. I don't know if George Green still offers it but there used to be a program named HedgeHog.

Why bet two horses to win? It certainly is not going to be popular with some bettors. The late Doc Sartin advocated betting multiple win bets. I believe he taught it more as a psychological ploy more than anything else. Going to the window to cash more often than you would only betting one horse can do wonders for your confidence.

If at the end of the day, you're kicking yourself with a lot of would've, should've, could've self talk, consider dutching two horses to win before doing anything more.

http://www.dmtc.com/handicapping/tools/wt_hedge.php

[cj's dad]

That's cold. So, how much did you win?

Read posts #14 & 17 - I know when I'm being called out - I wasn't born last night.

From my TVG account -

7/24/2009 12:39 PM DC0AD4C573C0 Monmouth Park 7 $5.00 WN 2 ONLINE $5.00 $0.00

7/24/2009 12:40 PM 380AEF0373C0 Monmouth Park 7 $4.00 WN 6 ONLINE $4.00 $131.60

My answer was in response to the OP. Yes, it can be wise to wager two in the same race to win, but not always. In this case it was.

[dutchboy]

http://www.dmtc.com/handicapping/tools/wt_dutch.php

[jonnielu]

Wow the truth comes out, Yes Sir.


Patrick

So, have you got something to say? Or, did you just stop in for a little quick name-calling?

jdl

[bisket]

lansdale when theres more than one overlay in a race. you'll find da bisket doing cartwheels. :jump:

[Irish Boy]

Thanks, you know, I've never stepped back to really consider how the results of horse races always make crystal clear sense before. Now I see the error of my ways.

I'm sorry, I forgot that you don't usually play horses that go of the board at better than 4-5. Everything else is just dice-throwing, right? A $2.60 place payout must be euphoria to you then. Making a cool $1.40 when your chalk horses come 1-2 is as good as it gets, no?

[GameTheory]

If the math is telling you that you are leaving money on the table by restricting yourself to one horse, then bet more than one. If you can, bet more than two! If there is a strong favorite, and you're sure he's going to lose -- bet the rest of the field! I think your goal should be to bet as many as you can get away with -- as many horses that give you an edge.

Now the math involved in making the "optimal" wager on multiple horses can get a bit complicated, but it is "never appropriate" to bet more than one horse? Nonsense. You are not "resigning yourself to lose one bet" (or more) -- the whole wager -- no matter how many horses are involved, should be seen as one bet, and constructed as such. If you walk away from the race with more than you put in, you've won. Now the payoffs are lower percentage wise, but the risk is much lower as well -- so you per-race profit simply is closer to your global profit. If you could win every race, but only make 1% profit per, would you do it? You'd be a fool not to.

If you want to bet pk3s and trifectas too, go ahead. But no need to stay out of the win pool with multiple picks.

[bisket]

Read posts #14 & 17 - I know when I'm being called out - I wasn't born last night.

From my TVG account -

7/24/2009 12:39 PM DC0AD4C573C0 Monmouth Park 7 $5.00 WN 2 ONLINE $5.00 $0.00

7/24/2009 12:40 PM 380AEF0373C0 Monmouth Park 7 $4.00 WN 6 ONLINE $4.00 $131.60

My answer was in response to the OP. Yes, it can be wise to wager two in the same race to win, but not always. In this case it was.
those betting amopunts mirror mine. the 4$ would be my raise in the last 2-4 minutes, but if the odds were 30-1 my raise probably would have been 20$

[jonnielu]

Read posts #14 & 17 - I know when I'm being called out - I wasn't born last night.

From my TVG account -

7/24/2009 12:39 PM DC0AD4C573C0 Monmouth Park 7 $5.00 WN 2 ONLINE $5.00 $0.00

7/24/2009 12:40 PM 380AEF0373C0 Monmouth Park 7 $4.00 WN 6 ONLINE $4.00 $131.60

My answer was in response to the OP. Yes, it can be wise to wager two in the same race to win, but not always. In this case it was.

So, how come I don't get no credit for showing you that documentation trick?

jdl

[pktruckdriver]

http://www.dmtc.com/handicapping/tools/wt_dutch.php


Thanks Dutch, this is better than the one I had, and much easier to use.

I been Win Dutch betting for awhile now and with the right odds it can be a good bet, but not all races allow for it, but when the race does , then you need to hammer it.

Name calling, huh, sorry Jon , the truth hurts, but no name calling here...


Patrick

[bisket]

If the math is telling you that you are leaving money on the table by restricting yourself to one horse, then bet more than one. If you can, bet more than two! If there is a strong favorite, and you're sure he's going to lose -- bet the rest of the field! I think your goal should be to bet as many as you can get away with -- as many horses that give you an edge.

Now the math involved in making the "optimal" wager on multiple horses can get a bit complicated, but it is "never appropriate" to bet more than one horse? Nonsense. You are not "resigning yourself to lose one bet" (or more) -- the whole wager -- no matter how many horses are involved, should be seen as one bet, and constructed as such. If you walk away from the race with more than you put in, you've won. Now the payoffs are lower percentage wise, but the risk is much lower as well -- so you per-race profit simply is closer to your global profit. If you could win every race, but only make 1% profit per, would you do it? You'd be a fool not to.

If you want to bet pk3s and trifectas too, go ahead. But no need to stay out of the win pool with multiple picks.
if you've got that much confidence in two horses than its only logical to play an exotic. heres what a tri would look like
a,b/a,b/c,d,e

[cj's dad]

those betting amopunts mirror mine. the 4$ would be my raise in the last 2-4 minutes, but if the odds were 30-1 my raise probably would have been 20$

2 points -

#1 - notice the time I posted the bets - I wagered early then had errands to run and didn't know I had won MP's 7th until later.
#2 - I normally bet a minimum of $5 to win - the $4 wager was a slip of the mouse.

[jonnielu]

I'm sorry, I forgot that you don't usually play horses that go of the board at better than 4-5. Everything else is just dice-throwing, right? A $2.60 place payout must be euphoria to you then. Making a cool $1.40 when your chalk horses come 1-2 is as good as it gets, no?



It amazes me that someone is actually willing to display such a lack of knowledge out in public. :eek:

jdl

[ryesteve]

Why would you bet two horses to place? That makes almost no senseI'd have taken out the word "amost" even before noting the source of that advice.

[bisket]

2 points -

#1 - notice the time I posted the bets - I wagered early then had errands to run and didn't know I had won MP's 7th until later.
#2 - I normally bet a minimum of $5 to win - the $4 wager was a slip of the mouse.
that only happens when you win :bang:

[GameTheory]

if you've got that much confidence in two horses than its only logical to play an exotic. heres what a tri would look like
a,b/a,b/c,d,eGo ahead and play that. And if you've only got a few bucks to put on the race, then you'll have to make a choice as to where you're going to put it. But no reason not to play in the win pool too if you want to spread it around. Just a matter of the risk/reward profile you're comfortable with. I'm just reacting to the idea that you're somehow screwing yourself to bet multiples in the win pool as opposed to forcing yourself to narrow it down to one when in fact it is the smart thing to do when you've got an edge on multiple horses.

[dutchboy]

http://en.wikipedia.org/wiki/Dutch_Schultz

Scroll down to section titled "The Numbers Game"

The term dutching is rumored to be named after the gangster Dutch Schultz.

[DJofSD]

if you've got that much confidence in two horses than its only logical to play an exotic. heres what a tri would look like
a,b/a,b/c,d,e
Say what? Only logical?

If you believe that the place horse is the 2nd best win horse, you might cash the ticket. But if the "other" win bet runs third or off the board, what do you have? Looks like nada to me.

You've got spreaders disease. The thread is about betting two horses to win. Not using multiple winners on top of verticle wagers.

[positive4th]

My 2 cents...........if you're using a Dutch calculator to isolate 2 or 3 horses and come away with a pre-determined profit, you are conceding that you have 1 basic opinion, and that is merely that you don't think the favorite can win (if you're Dutching WITH the favorite in the mix, then GL to you).

If you don't think the favorite can Win, why not simply play small Trifecta wheels that don't allow him to WIN but force him to finish in the money - - seems like a better chance to make $$ on your opinion than multiple win bets. So say that beatable fave is "A" and your choices are "K1 & K2".......along with some others who might have a chance, why not try something like this:

K1,K2 / A / K1,K2,B,C,D,E ($10)

K1,K2 / K1,K2,B,C,D,E / A ($10)

Now for $20 you're looking at a pretty decent cashout if either of your picks wins, and presumably more than if you'd simply bet to Win on both. If the favorite wins, you really haven't lost anything b/c you were trying to beat him anyhow

[positive4th]

My 2 cents...........if you're using a Dutch calculator to isolate 2 or 3 horses and come away with a pre-determined profit, you are conceding that you have 1 basic opinion, and that is merely that you don't think the favorite can win (if you're Dutching WITH the favorite in the mix, then GL to you).

If you don't think the favorite can Win, why not simply play small Trifecta wheels that don't allow him to WIN but force him to finish in the money - - seems like a better chance to make $$ on your opinion than multiple win bets. So say that beatable fave is "A" and your choices are "K1 & K2".......along with some others who might have a chance, why not try something like this:

K1,K2 / A / K1,K2,B,C,D,E ($10)

K1,K2 / K1,K2,B,C,D,E / A ($10)

Now for $20 you're looking at a pretty decent cashout if either of your picks wins, and presumably more than if you'd simply bet to Win on both. If the favorite wins, you really haven't lost anything b/c you were trying to beat him anyhow

And before someone kicks this, you can go ahead and bet small on K1 & K2 to win if you must

[positive4th]

if you've got that much confidence in two horses than its only logical to play an exotic. heres what a tri would look like
a,b/a,b/c,d,e

As a Tri-Wheeler myself, I am personally offended by this suggestion :) Please take that with some intended humor, but also look seriously at what is wrong here......you might like 2 horses to WIN, but that is far different than liking the same 2 to run FIRST & SECOND................that's why I'm always an advocate of hooking up your bright ideas (value horses) and then singling the "logical" horse underneath.........

[Overlay]

I only know what I've read on the board or heard about the Sartin methodology, but isn't betting two horses to win standard practice there?

[ezrabrooks]

My 2 cents...........if you're using a Dutch calculator to isolate 2 or 3 horses and come away with a pre-determined profit, you are conceding that you have 1 basic opinion, and that is merely that you don't think the favorite can win (if you're Dutching WITH the favorite in the mix, then GL to you).

If you don't think the favorite can Win, why not simply play small Trifecta wheels that don't allow him to WIN but force him to finish in the money - - seems like a better chance to make $$ on your opinion than multiple win bets. So say that beatable fave is "A" and your choices are "K1 & K2".......along with some others who might have a chance, why not try something like this:

K1,K2 / A / K1,K2,B,C,D,E ($10)

K1,K2 / K1,K2,B,C,D,E / A ($10)

Now for $20 you're looking at a pretty decent cashout if either of your picks wins, and presumably more than if you'd simply bet to Win on both. If the favorite wins, you really haven't lost anything b/c you were trying to beat him anyhow

"(if you're Dutching WITH the favorite in the mix, then GL to you)."

Have you ever Dutched the Favorite to 'break even', with profit on the another runner(s)?

Ez

[ryesteve]

Have you ever Dutched the Favorite to 'break even', with profit on the another runner(s)Might be good psychologically, but it's not good for your bankroll.

[ezrabrooks]

Might be good psychologically, but it's not good for your bankroll.

Thanks for the heads up..

Ez

[positive4th]

"(if you're Dutching WITH the favorite in the mix, then GL to you)."

Have you ever Dutched the Favorite to 'break even', with profit on the another runner(s)?

Ez

No, sure haven't

[bisket]

dj and others. I ALWAYS HAVE A WIN BET ON EVERY RACE. the tri is to increase my options to make a good hit. if i'm playing two win bets you can betcha i've found a race that has two overlays in it, and thats the type of race i hit hard. now one point if c,d, and e are the fav i just increase my wager amount and play an exacta. there is absolutely no reason to play a tri with the fav in the three spot. it won't pay much more than an exacta will pay.

[fmolf]

dutching like this all the time will probably shorten any losing streaks.Will definetly increase your hit rate .....but you will be probably trading awat some of your roi becaus as one poster already said you are depressing your odds by doing this.I am still of the opinion that if your handicapping is good you should bet your top overlayed selection as a prime bet.

[CBedo]

My simple minded question of the day is?

What difference does it make if you bet two horses in one race at good odds or bet one horse in two races at good odds.Either strategy could be correct depending on the situation, but the answer to your question is betting two horses in one race guarantees at least one losing bet thus reducing your effective "good" odds.

[GameTheory]

Might be good psychologically, but it's not good for your bankroll.Including the favorite in a multiple horse win wager is sometimes also the mathematically correct play. The lower the odds on the favorite, the more likely this is to be the case, even if by itself the bet on the favorite has a (slightly) negative expectation. This has to do with capital preservation and maximizing bankroll growth. What you don't lose this race you still have to bet with next race...

[PaceAdvantage]

My simple minded question of the day is?

What difference does it make if you bet two horses in one race at good odds or bet one horse in two races at good odds.NONE. Well said!

Why should one abandon the concept of the overlay in any situation?

[Light]

I think this real life example will demonstrate the value of betting 2 horses. This was my handicapping for today's pk6 at Dmr:

7/25/09
Top pick 2nd choice

Dmr 5th 4 2nd 5 $16.40
Dmr 6th 6 3rd 8 4th
Dmr 7th 10 $5.20 4 4th
Dmr 8th 1 5th 3 $30.20
Dmr 9th 2 3rd 7 $10.00
Dmr 10th 10 6th 14 $6.80
Return $5.20 Return $63.40



A no brainer. Is it wiser to invest $12 and get back $5.20 or invest $24 and get back $68.60. Also its our second choice that usually has the value horse.

[ryesteve]

Including the favorite in a multiple horse win wager is sometimes also the mathematically correct play even if by itself the bet on the favorite has a (slightly) negative expectation. This has to do with capital preservation and maximizing bankroll growth
I'm not saying you're wrong, but the idea of negative expectation bets maximizing bankroll growth is a tough one. Definitely something to think about...

[DanG]

When is it appropriate?
Coach;

An overlay is an overlay is an overlay. It can occur when your view says 3/5 and the public says 7/5 and / or when you view the favorite as vulnerable and your line says 3/1 & 4/1 and the public says 9/1 & 12/1.

In the interests of saving bandwidth and my up too late brain cells. :eek: “GameTheory” is a poster in this thread and (imho) you would be well served reading his thoughts and searching his past posts on this subject and more.

I don’t know if it’s the altitude in Colorado or what, but if you’re picking PA members for a new betting syndicate; “GameTheory” is on the very short list. (No disrespect meant to the many other great minds who post here btw.)

[JohnGalt1]

I think this real life example will demonstrate the value of betting 2 horses. This was my handicapping for today's pk6 at Dmr:

7/25/09
Top pick 2nd choice

Dmr 5th 4 2nd 5 $16.40
Dmr 6th 6 3rd 8 4th
Dmr 7th 10 $5.20 4 4th
Dmr 8th 1 5th 3 $30.20
Dmr 9th 2 3rd 7 $10.00
Dmr 10th 10 6th 14 $6.80
Return $5.20 Return $63.40



A no brainer. Is it wiser to invest $12 and get back $5.20 or invest $24 and get back $68.60. Also its our second choice that usually has the value horse.


Macho handicappers find one horse. Smart handicappers get the money.

Congrats.

Did you play and win the pick 4 with 2/2/2/2 ticket?

[ryesteve]

I'm not saying you're wrong, but the idea of negative expectation bets maximizing bankroll growth is a tough one. Definitely something to think about...
Ok, now I can say I disagree. I've run some simulations, and I see these negative expectation hedge bets on the favorites negatively impacting bankroll. Perhaps the results are contingent on the assumptions and parameters used, but the differences I'm seeing are pretty stark. Here's what I looked at:

Main bets are 4/1, ROI of +10%. Bet size = 1% of bankroll. Favorite odds are 8/5, ROI of -10%. Bet size = enough to cover bet size of main play.

I ran trials of 1000 plays, and then repeated those 1000 play trials 10,000 times. What I found is that without hedge betting, 8,995 of these 1000 play trials ended up showing a profit, with an average bankroll growth of 171%. With hedge betting, only 6171 trials showed a profit, and the average bankroll growth was only +44%.

Assuming higher odds on your main plays (hence more volatility) doesn't really change things. If I assume 8/1 odds instead of 4/1 odds, 7392 trials without hedge betting showed a profit, and the average bankroll growth was also 171%. With hedge betting, 4871 trials were profitable and again the average bankroll growth was 44%. In this instance, hedge betting also increased the chances of tapping out (defined as an ending bankroll under 10% of the starting bankroll). Without hedge betting there were 5 tapouts, and with hedge betting there were 26 (using 4/1 as the odds assumption, there were no tapouts either way)

So, the way I've looked at things here, these hedge bets are killing your bankroll the same way it kills your ROI. Perhaps there's a set of circumstances where this isn't the case, but so far, I'm not seeing it.

[JohnGalt1]

I had a debate with a friend who bets 3 horse exacta boxes in every race, who questioned me on races where I bet two to win with the logic that even if one horse wins I'd have one losing bet.

I don't think he understood that if he wins the exacta he would have 5 losing bets.

[CincyHorseplayer]

This isn't about philosophy or beliefs.If there is a logical way to make more money betting in any manner,that's the right approach.

I think far too many are married to the comfort of their betting methods and ponder the positive ROI concept as if it were the riddle of the Sphinx!

If you are unwilling to alter your view of bets to win more money,does that make you right or wrong or obstinately worse than wrong???;)

[Irish Boy]

:1: 10-1 (15%) for $100
:2: 12-1 (10%) for $100

Assuming the bettors assumptions are accurate, betting the 1 horse should net you $1600 over 20 races. Betting the 2 horse should get you $800. Betting the two together in each race should get you... $2400 (for $200 bet in each race, so the ROI is the same). I was skeptical when I started this, since I was convinced that you'd be eating away whatever advantage you had by betting both since you could only win once. But it doesn't seem to matter.

[dutchboy]

Del Mar Race 8 the Eddie Read

Based on the computer program the choice was to bet Thorn Song and or Global Hunter. For some reason I chose to only bet Thorn Song who decided to take a right turn on a left turn track at the wrong time. Global Hunter won and paid 30.20

So that is a reason why it may be a good reason to bet more than one horse in a race. Woe is me for a while.

[cj's dad]

So, how come I don't get no credit for showing you that documentation trick?

jdl

I'm pretty sure it's known as "copy and paste"

[Light]

Macho handicappers find one horse. Smart handicappers get the money.

Congrats.

Did you play and win the pick 4 with 2/2/2/2 ticket?

No.My focus was to hit the pk6.It was only afterwards that I realized what you said about the pk4 and did a Homer Simpson: "DOH!". 5 of 6 paid almost as much as the pk4 but I only got 4 of 6 after losing the 2nd leg and singling the chalk in the Eddie Read on a $32 ticket.

[Red Knave]

Perhaps the results are contingent on the assumptions and parameters used, but the differences I'm seeing are pretty stark. Here's what I looked at:Steve, GT said slightly negative. I think -10% is more than "slightly" negative.

Main bets are 4/1, ROI of +10%. Bet size = 1% of bankroll. Favorite odds are 8/5, ROI of -10%. Bet size = enough to cover bet size of main playSo, you are betting 1% of bankroll and dutching to the odds but assuming that the 8/5 horse wins 10% less than he should? Or wins as often as it should but wins you 10% less than it should?

I ask because this because, to me, this doesn't make sense.
With hedge betting, only 6171 trials showed a profit, and the average bankroll growth was only +44%.I would have expected more trials to make a profit (or at least a similar %) but with less bankroll increase and also fewer tapouts.

Thanks for testing this theory by the way.

[CBedo]

:1: 10-1 (15%) for $100
:2: 12-1 (10%) for $100

Assuming the bettors assumptions are accurate, betting the 1 horse should net you $1600 over 20 races. Betting the 2 horse should get you $800. Betting the two together in each race should get you... $2400 (for $200 bet in each race, so the ROI is the same). I was skeptical when I started this, since I was convinced that you'd be eating away whatever advantage you had by betting both since you could only win once. But it doesn't seem to matter.???? Getting 10/1 with a 15% win probability is a 65% edge. That's 1300 expected value over 20 bets of $100. 10% win probability at 12/1 over 20 bets of $100 is a $600 expectation.

And yes, betting $100 on both per race will yield more total dollars, but now you're comparing betting $200 to betting $100. So the ROI is not the same. You are investing twice as much. The gross expectation is the sum of the two (thus more), but the ROI is less.

[Irish Boy]

You'll need $2,000 to make 20 $100 bets. If you hit three of them (15%) at 10-1, you'll make $3300 back (the $1000 you won and the $100 you bet). Subtract the $1,700 you'll lose and that leaves you at $1,600. You can take out the $300 you bet on winning wagers as well, but as long as you take that money out of the equation everywhere the results will come out the same.

The ROI is only less if you were always betting the 1 and never the two, since I made the edge larger in the one. If you have a horse that has a larger edge, you should always just bet that one. Fair enough. But if you have to horses with identical edges (suppose two 10-1 horses you think will win 15% of the time), the ROI will be the same regardless of whether you consistently bet one or if you bet both. Betting both the 1 and 2 gives you the ROI for those horses averaged.

[CBedo]

You'll need $2,000 to make 20 $100 bets. If you hit three of them (15%) at 10-1, you'll make $3300 back (the $1000 you won and the $100 you bet). Subtract the $1,700 you'll lose and that leaves you at $1,600. You can take out the $300 you bet on winning wagers as well, but as long as you take that money out of the equation everywhere the results will come out the same.

The ROI is only less if you were always betting the 1 and never the two, since I made the edge larger in the one. If you have a horse that has a larger edge, you should always just bet that one. Fair enough. But if you have to horses with identical edges (suppose two 10-1 horses you think will win 15% of the time), the ROI will be the same regardless of whether you consistently bet one or if you bet both. Betting both the 1 and 2 gives you the ROI for those horses averaged.1) You shouldn't count the $300 you bet as part of your return, so you'll make 3000 on the 3 winners and lose 1700 on the other 17 for a $1300 profit (if you're going to judge by total return, you should bet more! haha).

2) The ROI is less. Betting two 10/1 shots in the same race (for example), will equate to a 5/1 return since you are betting twice as much.

[GameTheory]

Ok, now I can say I disagree. I've run some simulations, and I see these negative expectation hedge bets on the favorites negatively impacting bankroll. Perhaps the results are contingent on the assumptions and parameters used, but the differences I'm seeing are pretty stark. Yes, the results are contingent. This is a fairly rare technical exception -- I'm not suggesting everyone should hedge bet on favorites willy-nilly. It also only really comes up in an optimal (Kelly) betting framework, which is the lens that ALL wagering strategies should be analyzed with to find their objective worth, even if you aren't actually Kelly betting.



Here's what I looked at:

Main bets are 4/1, ROI of +10%. Bet size = 1% of bankroll. Favorite odds are 8/5, ROI of -10%. Bet size = enough to cover bet size of main play.
The situation only comes up with STRONG favorites -- 8/5 is not strong enough -- even money and below what we're looking for. And it also has to be a legitimate strong favorite, deserving of those odds. The "negative expectation" would be no more than 2%-3% in order for it to make sense. In short, a race in which the favorite is VERY likely to win, but in which you nevertheless have another horse going off at high odds that is an overlay. In cases like that, optimal wagering will indicate a bet on both horses even though the favorite is not a true overlay because the most likely (overwhelmingly so) scenario is for the favorite to win and making that bet minimizes your losses. Bet sizing is also important here -- I'd have to work up a real example with numbers to show you. Maybe I'll do that later. And it is also assuming you are using Kelly type betting, which uses a percentage of bankroll, so saving some capital from this play gives us more to work with next play, which is why it is about bankroll growth rather than hedging this single play. These are good situations -- normally when you've got a strong high odds overlay you're still not very likely to win. So in these cases where making the hedge is actually the smart play, you're going to at least cash a ticket 70% of the time.



I ran trials of 1000 plays, and then repeated those 1000 play trials 10,000 times. What I found is that without hedge betting, 8,995 of these 1000 play trials ended up showing a profit, with an average bankroll growth of 171%. With hedge betting, only 6171 trials showed a profit, and the average bankroll growth was only +44%.

Assuming higher odds on your main plays (hence more volatility) doesn't really change things. If I assume 8/1 odds instead of 4/1 odds, 7392 trials without hedge betting showed a profit, and the average bankroll growth was also 171%. With hedge betting, 4871 trials were profitable and again the average bankroll growth was 44%. In this instance, hedge betting also increased the chances of tapping out (defined as an ending bankroll under 10% of the starting bankroll). Without hedge betting there were 5 tapouts, and with hedge betting there were 26 (using 4/1 as the odds assumption, there were no tapouts either way)

So, the way I've looked at things here, these hedge bets are killing your bankroll the same way it kills your ROI. Perhaps there's a set of circumstances where this isn't the case, but so far, I'm not seeing it.And people should take what you are saying here to heart as well -- betting on favorites just for the hell of it is a bankroll killer. I was just pointing out a mathematical exception. The problem is since almost no one really understands optimal betting, and you have to in order to truly understand (technically, anyway) ANY betting, when it comes to devising wagering strategies, most people are at worst just guessing (maybe a hedge is a good idea here?) to somewhat better trial-and-error until good results are seen to best empirical methods like the simulations you ran above.

But there are in fact formulas that will tell you if any bet is good, and how good it is. Unfortunately, when it comes to multiple horse betting, those formulas become non-linear and you can no longer just plug-in the numbers -- you've now got to go through a iterative process to arrive at the correct answers. A process that is programmed into exactly zero pieces of gambling software (horse racing or otherwise), at least that I've ever heard of, which shows just how rare the understanding of it is. (Most discussion of Kelly-type wagering, even in fairly learned books, is completely wrong, by the way -- or wrong enough to give you the well, "wrong" idea.)

[ryesteve]

Steve, GT said slightly negative. I think -10% is more than "slightly" negative.Yes, that's pretty arbitary. And I was thinking in terms of someone using this as a standard strategy. I see now that GameTheory was speaking in far more selective terms, but I got the impression some in this thread were not, and I know in the past, some people here have advocate negative expectation hedge bets as an everyday strategy, so the -10% seemed like a fair estimate of what one could expect with frequent bets on favorites.

So, you are betting 1% of bankroll and dutching to the odds but assuming that the 8/5 horse wins 10% less than he should? Or wins as often as it should but wins you 10% less than it should?.I'm not sure what you mean by "should" here, so i'll just explain what I did... I set the win% such that the 8/5 payout would generate a long-term ROI of -10%.

I would have expected more trials to make a profit (or at least a similar %) but with less bankroll increase and also fewer tapouts.Honestly, I didn't expect the difference to be this wide either, which is why I took the time to test it. But in hindsight, I suppose it makes sense that your bankroll is being drained by negative expectation bets. You might feel as if you're preserving bankroll when the favorite wins, but over time, since you're only getting back 90 cents for edge hedge dollar bet, I can see how it can erode your bankroll.

This is a fairly rare technical exception. I was just pointing out a mathematical exception.Gotcha. Still, it was interesting to see what happens when this strategy is taken beyond the limits you've recommended, since I know there are some people out there betting this way, and I didn't expect the results to look as bad as they did.

[Irish Boy]

1) You shouldn't count the $300 you bet as part of your return, so you'll make 3000 on the 3 winners and lose 1700 on the other 17 for a $1300 profit (if you're going to judge by total return, you should bet more! haha).

2) The ROI is less. Betting two 10/1 shots in the same race (for example), will equate to a 5/1 return since you are betting twice as much.

Yes, but you'd be winning twice as often as well. Wining 10% of the time at 10-1 and 20% of the time at 5-1 is the same thing.

[GameTheory]

Ahh...another interesting question people struggle with. Once we decide that it is ok to bet two horses, the question is how much to bet? If we are a $20 flat bettor do we just double up and bet $40 on this race with two horses, or do we now only bet $10 each and keep our per-race amount the same?

[Red Knave]

I'm not sure what you mean by "should" here, so i'll just explain what I did... I set the win% such that the 8/5 payout would generate a long-term ROI of -10%. Okay, thanks. I didn't really think through the question.
What I meant by "should" is that horses with final odds of 8/5 win about 31.5% of the time. In the test they would have to win more often to compensate for the take and only lose 10% (i.e. they would need to win about 34.5%) as would the 4/1 horses to get the +10% ROI. I guess I made the assumption that your line would have to be, say, 8/5 and 5/2 in order to bet both horses at all.
As GT alludes to, and as you probably know, odds on horses are typically overlays enough to cover at least part of the take. Any pre-race estimate of expectation should be closer to reality for your low odds horse than for your longer odds horse. If your 2 horses are 3/5 and 12/1 and you have them on your line as 4/5 and 4/1 which reality do you believe?

[CincyHorseplayer]

Somebody just said it.While you will be betting more,you're hit rate will go up as well.The Sartin theory on Dutch betting was achieving a 50-60% strike rate.That can work just as well as a 35% strike rate with one horse and it has been posted here many times,how much more favorites are winning.To achieve that 35% it could be argued that the avg mutuel will be less on a single win bet than a double.

[ezrabrooks]

Ahh...another interesting question people struggle with. Once we decide that it is ok to bet two horses, the question is how much to bet? If we are a $20 flat bettor do we just double up and bet $40 on this race with two horses, or do we now only bet $10 each and keep our per-race amount the same?

I am not a Kelly bettor...so I just use my flat amount on each. When I am covering a short price fav, then I dutch to break even on the fav.

Ez

[CBedo]

Yes, but you'd be winning twice as often as well. Wining 10% of the time at 10-1 and 20% of the time at 5-1 is the same thing.:bang: I was brain dead on that one.

[Irish Boy]

:bang: I was brain dead on that one.

No problem. I actually wrote out a whole post preporting to show how win betting on two horses was almost always a losing proposition... and then I actually did the math. Oops.

[Overlay]

Yes, but you'd be winning twice as often as well. Wining 10% of the time at 10-1 and 20% of the time at 5-1 is the same thing.

Technically, wouldn't winning 10% of the time at 10-1 produce a minimum $NET of $2.20 per $2.00 wager, whereas winning 20% of the time at 5-1 would yield a minimum $NET of $2.40 per $2.00 wager?

[Irish Boy]

Technically, wouldn't winning 10% of the time at 10-1 produce a minimum $NET of $2.20 per $2.00 wager, whereas winning 20% of the time at 5-1 would yield a minimum $NET of $2.40 per $2.00 wager?

I bet $10 on Gluestick Charlie at 10-1... payout is $22.00. I get $110.00

I bet $20 on Old Elmer at 5-1... payout is 12.00. I get $120.00

Hmmm.... seems you're right. Need to think this over.

[kenwoodallpromos]

Sorry to oversimplify- If $4.00 in bets per race return you an average of over $4.00 per race- you win! I believe combined odds multiplied by combined win %= return.
If an yone has stats on favorite win % by type of race (sprints on each surface, and turf routes; good and fast tracks), please post.

[ryesteve]

Hmmm.... seems you're right. Need to think this over.The problem is the original premise that betting two horses at 10/1 is the same as betting one at 5/1, was not correct. It's the same as betting one at 9/2. You can't just cut the "10" in half, because "10/1" is not "1 out of 10", it's "1 out of 11". So if you double that, you have 2 out of 11, which is the same as 9/2.

[Irish Boy]

The problem is the original premise that betting two horses at 10/1 is the same as betting one at 5/1, was not correct. It's the same as betting one at 9/2. You can't just cut the "10" in half, because "10/1" is not "1 out of 10", it's "1 out of 11". So if you double that, you have 2 out of 11, which is the same as 9/2.

Ah yes. Thank you, that's right.

[jonnielu]

I'm pretty sure it's known as "copy and paste"

So how come when I do it, I gotta have 3 clergymen for witnesses besides?

jdl

[jonnielu]

Thanks Dutch, this is better than the one I had, and much easier to use.

I been Win Dutch betting for awhile now and with the right odds it can be a good bet, but not all races allow for it, but when the race does , then you need to hammer it.

Name calling, huh, sorry Jon , the truth hurts, but no name calling here...


Patrick

What truth is that? I'm staying open late tonight just to accept your apologies.

jdl

[PaceAdvantage]

So how come when I do it, I gotta have 3 clergymen for witnesses besides?

jdlSounds like some sort of trust issue to me...call me crazy!

[jonnielu]

Sounds like some sort of trust issue to me...call me crazy!

What's not to trust? This is my question, what would I have to do?

jdl

[PaceAdvantage]

What's not to trust? This is my question, what would I have to do?

jdlYou tell me...you're the one that wrote that you feel you need 3 clergymen as witnesses while others seem to not require such things before people take their word...