An odds cap on your overlay

[porchy44]

Is anyone else confronted with too much of an overlay at times. There are times I would like my selection to go off no higher than say 4/1 or 5/1. .If he climbs to say 9/1 . I take this to mean that I have not analyzed the race properly. My records indicate a far lower win rate when a horse goes off at a higher odds then my cap I have placed on him.

I guess Im sayiing is If I perceive my pick to be an overlay at 3/1, I do not like him as much if he goes off at 6/1.

Just wondering if anyone else looks at overlays this way.

[statik27]

It usually doesn't bother me if I see my horse floating up, but it depends on the track and that tracks betting population. It could just be that people are jumping on a short odds favorite thats taking alot of steam and its distorting the odds of your choice. In which case its a good thing.


statik

[bigmack]

Ain't no such thing as too much of an overlay. I don't respect mob mentality enough to talk myself off what I see as an overlay cause they can't see what I might or they have interests in a false fav

[Premier Turf Club]

In theory, I agree there can't be "too much of an overlay" but over the course of thousands of events Benter and others have detremined that the "best" odds-line will be about 55% your line (assuming you make a good line) and 45% the public's line. Again, this is over a large number of trials and doesn't represent what can occur on a single event.

The more the reasons for liking your horse are "hidden" (recent unreported work, new medication, 1st time gelding, bad trip last out not obvious to those reading the form, etc.) the more likely an overlay you're getting is a "good overlay." The more the horse you like is fully exposed, the more likely it is a false overlay, the result of something negative you can't readily see (e.g. horse with fastest most recent speed figures is 2-1 in the ML but is 4-1 with 1 MTP because it turns out he was coughing yesterday).

Unfortunately, no such thing as a free lunch (although with some ADWs enough rewards points might get you a meal at McDonalds;)).

[spicytomato]

i seen some horses come in first at 80/1 also 99/1 :eek:
just catch it becuz everything gets quiet when they
stare at the race when its happening

now thems bets i wish i had wagered:sleeping:

happens:faint:

[Overlay]

I've seen too many instances of the public overbetting a particular angle or factor (like a hot jockey), with the effect of driving up the odds of other live horses in the race and making them overlays. (Not that the factor the public is jumping on doesn't have some significance, but just not to the extent that the public believes.) I think it also helps if you can find plausible reasons why other horses in the race are being overbet, or why your selection is being underbet, and discount them. And record-keeping can help build confidence in situations like that, too.

[Zman179]

With all due respect Porchy, that has got to be one of the most asinine and self-destructive comments that I've ever heard in my 25 years of betting the nags.

Why do you care what someone else's opinion on a race is? Maybe you spotted a live horse that nobody else spotted? What you said was equivalent to saying, "I'm only going to allow myself to win X amount of dollars and no more." That 25/1 shot might be the one that pays your rent for a couple months, or allows you and your significant other to take a free vacation to the tropics, or allows you to put a down payment on that nice car that you've wanted. Yet you would deny yourself this possibility because of what the public thinks.

If I were you, if you really like a horse, and he's 25/1, BET WITH BOTH FISTS. Why hit for a bunt when you can smack one out of the park?

[098poi]

I have learned to respect the public's opinion. I have gone through phases of betting too many high priced horses thinking I had an overlay or just trying to beat the favorite. I end up losing money. If one is selective and has multiple reasons to predict an overlay (as a previous poster said) then the higher the odds the better. It comes back to basics though. It depends on the individual and his own past performance. If you keep records of how often your 2 to 1's hit and your 3 to 1's hit etc., then you'll know if a certain scenario is a true overlay. (As said by Barry Meadow) I don't keep the best records but one time I did a little survey from my Youbet wager log and noticed that my good win hits seemed to peak out at 6 or 7 to 1. I hit above that occasionally but probably not enough or with the consistency to justify a win bet. Alas I don't take my own advice. :bang:

[Tom]

Actually, Pizzola used that very idea at a seminar I attended in Philly a few years ago. He use three oddslines - the ML, an expected line, and his value line. The expected line was a reality check, just as Porchy talks about here. We even got a free program that calculated the expected odds line using Beyer numbers - called B Odds ( funny name choice, since the Hat Bradshaw had just come out with is A Odds!

But the idea was that if your horses wqas expected to be say 8-1 and it was really 50-1, that was a red falg.

Porchy, I will see if I can find my notes/taps of that and post more for you here. Or maybe Dick Schmidt remembers it - he co-hosted that seminar with Mikey. dick - the one in November, where the Eagles beat up the Rams during our day at the races. ;)

Or you could try this forum....do a search there, or join and post the Q. Some of the guys there were at the seminar, too.

http://www.sportsbettingacumen.com/forum/forum.asp?FORUM_ID=10

[098poi]

For those of you obsessed with winning, remember this: at the trap, it is the second mouse that gets the cheese.


I like that!

[Edward DeVere]

With all due respect Porchy, that has got to be one of the most asinine and self-destructive comments that I've ever heard in my 25 years of betting ?

Since he apparently has the records to back up what he's saying in regards to his own betting (not YOUR betting - HIS betting), it would appear that your comment is the one that's asinine.

[Zman179]

Since he apparently has the records to back up what he's saying in regards to his own betting (not YOUR betting - HIS betting), it would appear that your comment is the one that's asinine.

Whatever.

[The Judge]

If you truely believe that you are getting such a great overlay, that is you have a horse at 3/1 and the public has made such a huge error as to make the horse 30/1 then you should beg borrow and steal all the money you can get and bet it on that horse. Why ,because you know the public will not make that type of a mistake often.

No, we all know that if we have a horse at 3/1 and its 30/1 on the board we have made a mistake or the public has really missed something big time. What could they have missed? Is it in the racing form. Is it a hidden pace number, a lowly shipper from Grants Pass, breeding, a trainer move? You must know the answer if not "you" have made a mistake.

This is why I bet the less or the same amount of money but not "more money". I'm not 100 % sure but I must know why the public if so off the mark to bet any money. What I do expect is the 30/1 shot to run a lot better then 30/1 so yes I have my win ticket maybe a place and for sure the exotics.

[porchy44]

Tom -thanks for the link and in advance for any other information about the pizzola seminar.

098poi -interesting my records also show my best win hits also peak out at about 6 or 7/1.

Zman179- I know at first blush it seems "asinine" but what do you do when your records indicate different.

[pandy]

I would think you'd be better off actually betting more than your usual unit when the overlay is much higher than expected.

[PaceAdvantage]

If you truely believe that you are getting such a great overlay, that is you have a horse at 3/1 and the public has made such a huge error as to make the horse 30/1 then you should beg borrow and steal all the money you can get and bet it on that horse.

That's the worst bankroll management advice I think I've ever read here....all the horse has to do is trip and it doesn't much matter how legitimate your overlay was....you're now on the street.

[DeanT]

Interesting topic.

Seems to me if you have a horse at 3-1, and he is 10-1 on the board, you are betting an angle. The barn cash on the horse is dead, which may signal a red light. It happens.

Like most things in racing it is anomalous.

I had a friend tell me he had a horse called Running Book in a harness race last week at 5-1 on his line. He won by four, and won like a 4-5 shot. He paid $82.80. Go figure. Gotta love racing.

[bigmack]

The barn cash on the horse is dead, which may signal a red light. It happens.
All those in favor of dismissing much of the credence of "barn loot" raise your hand.

http://i165.photobucket.com/albums/u70/macktime/RaisedHands.jpg

It's official.

Barn dough takes a back seat to a jock change, a hot work, a last out troubled trip, med change, class drop, equip change, surf switch, distance adjust, pedigree aware world.

[cj]

Hand raised.

[JustRalph]

hand way up..............

I can think of a ton of Longshots that I watched drift way up.........while I was pushing money thru the windows on them. I agree with someone who mentioned that at some tracks this bothers me more than others.

I tend to play longshots that the public has just decided not to bet for reasons that are sometimes pretty obvious......

1. First on Turf.........
2. Early Speed that has faded before, going longer today
3. Big Droppers that look like a fire sale..........
4. Stakes Horses that have been beaten by the same crowd as today

I am usually going against the crowd on bets like these ...........wherein I have decided that my horse deserves a second chance. It is amazing to me how the crowd tends to never forgive "one bad race" If it's the last out especially.............this is why I value video replays so much...........

[cnollfan]

While the higher than expected odds may well mean your analysis is off base, the flip side is that you don't need to be right as often.

I am not bothered by my horse's odds drifting up and up, but if you are, one thing to look at might be what you think of the favorite. If you have sound reasons to dislike the favorite, heavy betting on that one that causes your horse's odds to rise should be welcomed, not feared.

Have you considered the psychological implications of passing a horse that you thought should pay $20 if it wins and pays $39.80? That could be devastating.

[The Judge]

not what can happen in a race. My statement may have been a bit too dramatic but my point is the same. If one felt his line was correct and the public "so incorrect" why not bet more money then normal "all the time" including those times when things don't go wrong. Shouldn't you make more money in the long run? Logic says yes, but in reallity you will find that the public makes some mistakes but not many as expressed or at lease not so many that you can feel comfortable taking the rubber band off the old bankroll everytime you "think" you have a hugh overlay.

The reason must people won't bet alot more money isn't because something can go wrong in the race ,that's the nature of horse racing thats the risk we accept in playing the game we accept that risk in every race we play. The reason I won't increase my bet is that I could be wrong in my assessment of the race. My 3/1 shot thats 30/1 on the board may not really be a 3/1 shot but a 9/1 shot . So what I expect to happen is for my horse to run better then the normal 30/1 shot. I will bet him but seldom will I bet more then normal.

If a person is "that" comfortable with their line making they should increase their bet as much as they can stand when their horses odds are at their highest compared to their line.

Thats my take on the problem

[levinmpa]

I like what Pizzola says in his seminars regarding "overlays" that are much higher than his value line. Ask yourself "why am I the lucky one". If odds are much higher than expected, it's a red flag. Of course some will win, but not enough to overcome the takeout.

[ryesteve]

If one felt his line was correct
But obviously it's not, based on the poster's experience. If he sees that horses that LOOK LIKE huge overlays against his line perform poorly, then his line is NOT correct in these instances. You people are all arguing that huge overlays should be bet aggressively, which really has nothing to do with the reality of the situation described. Based on his observed results, clearly his line is misevaluating these horses, so he would be correct in IGNORING his line and concluding that these horses are not actually overlays in the first place.

[betchatoo]

I don't let the odds line effect my wager except as a minimum. I base my wagers on strength of play with minimum odds (for example I may have a 3 unit play as long as I receive 4-1 or better) and the higher the odds, the better I like it. However, I do understand the hesitation if a horses odds are way above what you expect to get.

I remember when I first started betting football I followed a handicapper named Mike Lee. He made a weekly line and told you to follow it if you got 3 to 7 points better than the line he made. However, if the line was more than 10 points different, you were to back off, because it usually meant something he hadn't foreseen was happening. I don't follow that plan in my racing strategy, but I can see why someone might.

[porchy44]

But obviously it's not, based on the poster's experience. If he sees that horses that LOOK LIKE huge overlays against his line perform poorly, then his line is NOT correct in these instances. You people are all arguing that huge overlays should be bet aggressively, which really has nothing to do with the reality of the situation described. Based on his observed results, clearly his line is misevaluating these horses, so he would be correct in IGNORING his line and concluding that these horses are not actually overlays in the first place.


I think you nailed it in my situation.

Also levinmpa is the second person to mention the pizzolla seminar on overlays. I hope to find more info on this.

[The Judge]

does anyone else see overlays this way? I could have said yes or no but see nothing wrong with ansewering "how do I see overlays/ some overlays". I quess my real answer is it depends but normally its not enough to raise my bet because of my line. I will bet it though.

If you aren't willing to raise your bet, why? This is not a challenge I would like to know. The public has made what seems to be (in my senario) a hugh mistake according to our line do you take the bubber band off the bank roll or not. I don't.

[Pell Mell]

My question would be; do you figure the horse to win or are you betting just because it's an overlay? If you figure a horse should be 6/1, and he may actually be your third choice to win, are you betting it just because it's going off at 10/1? One could bet overlays from now till doom's day but if they don't figure to win what's the purpose? Being an overlay is not of itself going to make a horse win. I only bet horses that I figure to be best in the race and if the price is too short I just don't bet. In this day and age with all the tracks to choose from it's not that hard to find a solid play that's overlaid, one must have patience and they will turn up.

[ryesteve]

One could bet overlays from now till doom's day but if they don't figure to win what's the purpose?
What does "figure to win" mean? Every horse in a race has a probability of winning. Just because that probability of winning isn't the highest in the race has nothing to do with how profitable the bet may be. If one is able to assess those probabilties accurately, one can make a fortune while NEVER betting the horse with the best chance of winning.

Analogy: if I hand you two dice and tell you I'm offering 3-1 on a 7 and 10-1 on a 6 or an 8, are you still going to bet on 7 because it's the most likely outcome?

[DeanT]

All those in favor of dismissing much of the credence of "barn loot" raise your hand.

It's official.

Barn dough takes a back seat to a jock change, a hot work, a last out troubled trip, med change, class drop, equip change, surf switch, distance adjust, pedigree aware world.

Hi Bigmack,

In general I agree.

Barn's that bet in certain situations and so called smart money is something that I look at in instances, though. Recently for example, I got stung. There is a relatively new trainer. He is getting a few horses. Some are shipped to him. The first two won. At 8-5 to 3-1. I had them handicapped more like 8-1. Out of his next several they were live, hit the board, and were fairly well bet.

Then start nine or ten in this situation, the horse was dead on the board 11-1. I ignored the lack of cash and hamered him. Raced horrid. The next race off the terrible effort I chucked the horse, eventho he opened at 8-5, went up to 5-1 and dove to 3-1 last flash. The horse jogged.

I know we have seen the above many times- that is what I was alluding to in my post. It pays sometimes to pay attention to one or two stables at each track where barn cash is correlated to results, imo.

[Robert Fischer]

when a horse that should "obviously" be getting money is ice-cold on the board, then I will proceed with caution. A good example was Robe Decollete in the American Oaks. - It seemed to be public-obvious that the horse should be the top choice. Late Running style and American bias aside she should have recieved more money. With the Aussie horse recieving more attention than the seemingly superior Japanese animal it was time to proceed with caution or simply include the Aussie in the exotics.
Hard Spun was another example in the Preakness. While Hard Spun may have been the clear 2nd choice among track-goers, the "money" was strong for Curlin who went off as 2nd betting choice.

When there is a horse I like that doesn't have "public-obvious" positive factors (or he may have a prominent "false"-negative factor that betters will avoid), then I am clearly taking a stand against the public from the start and I expect the horse to be overlayed.

A third class is the low-probability-overlay, - for example I may think the horse has about a 5% chance of winning and he is 25 or 30-1 and I still may pass unless I am dealing with very small incremental bets.

[cj]

I'm not sure using races where the pools were huge, especially the Preakness, fits the bill here. Inside money would have little to no impact on those days. Those days are ripe for finding great overlays you won't on your run of the mill card.

[Scotchman]

I like what Pizzola says in his seminars regarding "overlays" that are much higher than his value line. Ask yourself "why am I the lucky one". If odds are much higher than expected, it's a red flag. Of course some will win, but not enough to overcome the takeout.

I once bought a win 4 ticket where I singled 2 horses that appeared to me that they were easily going to be the 2nd call leaders. I took All and All in the other 2 legs. I bought the ticket well in advance thinking that my singles would be paying in the $8 to $10 range. Well the first one won and paid $32.
This was in the days before live video, so I was getting the results on the phone. I phoned for the second leg of this bet, and to my amazement, the next single won and paid $47. My reward was a $15000 payout.

Had I been at the track, and seen the astronomical odds on my first single, I may not have pulled the trigger on the bet. Thankfully I wasn't there.

[Robert Fischer]

I'm not sure using races where the pools were huge, especially the Preakness, fits the bill here. Inside money would have little to no impact on those days. Those days are ripe for finding great overlays you won't on your run of the mill card.

The big pools are a different phenomenon to be sure.
While the barn isn't going to sway the odds as in a small race with little action, these races do attract some large wagers as well.

[Tom]

What does "figure to win" mean? Every horse in a race has a probability of winning. Just because that probability of winning isn't the highest in the race has nothing to do with how profitable the bet may be. If one is able to assess those probabilties accurately, one can make a fortune while NEVER betting the horse with the best chance of winning.

Analogy: if I hand you two dice and tell you I'm offering 3-1 on a 7 and 10-1 on a 6 or an 8, are you still going to bet on 7 because it's the most likely outcome?

You are more likely to get a 6 or 8 than a 7:

There are 3 ways to make a 7
1-6
2-5
3-4

There are 6 ways to make a 6 or 8

1-5
2-4
3-3

2-6
3-5
4-4

[SMOO]

Tom - I think he meant 6 OR 8, not both.

[1st time lasix]

Let's assume we have enough starters and there is some vulnerability among some of the low odd public favorites. From an exotics point of view....if I have three main overlay contenders in my opinion for the upcoming race: One at 5/2, one at 4/1 and the other at 13/1. I like to play the higher odds horse to win and then use him heavily as a key in the two and three hole in the exotics to push up the payoffs. Certainly double digit long shots win occasionally....but generally not enough to cover the losses. Now there are certain races where a longshot is more likely. That has been touched on with a previous post {maidens, first on grass, layoffs, etc}

[ryesteve]

Tom - I think he meant 6 OR 8, not both.
Yes, you're correct...

And Tom, there are 6 ways to make a 7, and 10 ways to make a 6 or 8...

[Tom]

How do you get 10? (5 my way - same thing, but halfed, same propabilities)

6 = 1 5, 5 1, 2 4, 4 2, 3 3 = 5 ways
8 = 2 6, 6 2, 3 5, 5 3, 4 4, = 5 ways

But you said 6 or 8, so you cannot count 5 of the ways.

7 = 1 6, 6 1, 2 5, 5 2, 3 4, 4 3 = 6 ways

[ryesteve]

That's correct... I was correcting your miscounting in your misinterpretation. You originally said "6 ways" which is not half of 10.

[Cratos]

Is anyone else confronted with too much of an overlay at times. There are times I would like my selection to go off no higher than say 4/1 or 5/1. .If he climbs to say 9/1 . I take this to mean that I have not analyzed the race properly. My records indicate a far lower win rate when a horse goes off at a higher odds then my cap I have placed on him.

I guess Im sayiing is If I perceive my pick to be an overlay at 3/1, I do not like him as much if he goes off at 6/1.

Just wondering if anyone else looks at overlays this way.

There are at least two ways (and probably many more) of assessing betting with respect to changing odds. If you believe that your horse should be 3-1 which is about a 25% chance of winning, but the betting crowd believes it should be 30-1 which is about a 3% chance of winning then ask yourself if your are 8 times better on this choice than the crowd. The reason being is sometimes we allow big odds to lure us into false wagers.

The second assessment can come from field size. The smaller the field size the greater the possibility that the odds established by the betting crowd are right.

[Overlay]

If you believe that your horse should be 3-1 which is about a 25% chance of winning, but the betting crowd believes it should be 30-1 which is about a 3% chance of winning then ask yourself if your are 8 times better on this choice than the crowd. The reason being is sometimes we allow big odds to lure us into false wagers.

Valid point, which speaks to the advantage of basing odds assessment on statistics rather than opinion. It allows you to see exactly why and how you assign any particular horse its chance of winning, and to retain confidence in your odds-assignment process (especially if, as I say, you can explain why a horse you like is being overlooked, why the public is overbetting one or more other horses, or both).

[Overlay]

My question would be; do you figure the horse to win or are you betting just because it's an overlay? If you figure a horse should be 6/1, and he may actually be your third choice to win, are you betting it just because it's going off at 10/1? One could bet overlays from now till doom's day but if they don't figure to win what's the purpose? Being an overlay is not of itself going to make a horse win. I only bet horses that I figure to be best in the race and if the price is too short I just don't bet. In this day and age with all the tracks to choose from it's not that hard to find a solid play that's overlaid, one must have patience and they will turn up.

Limiting your consideration only to the horse that you figure as the most likely winner can certainly be a valid way to play. But just as being an overlay doesn't mean that a horse will win, neither does being the best horse in the race (unless it's also the only horse in the race). Although it's easier for a lower-odds horse to be a greater proportional overlay, any horse can offer value based on the comparison of your assessment of its winning chances with its odds. It may not "figure" to win in the sense of being the "best" horse in the race. But it will win often enough that, when it does win, its overlaid odds will produce an overall profit. And betting with the percentages like that means that the more you bet, the greater the chance you're giving those favorable probabilities to assert themselves, rather than pursuing a wagering strategy that is more selective.

[dav4463]

I tried narrowing it down to one horse. What I found from my own records is that I missed a lot of second, third, or fourth choices that paid big odds. Some races I do narrow to one horse, but in a majority of races I can make a legitimate case for three or four horses.

[classhandicapper]

My question would be; do you figure the horse to win or are you betting just because it's an overlay? If you figure a horse should be 6/1, and he may actually be your third choice to win, are you betting it just because it's going off at 10/1? One could bet overlays from now till doom's day but if they don't figure to win what's the purpose? Being an overlay is not of itself going to make a horse win. I only bet horses that I figure to be best in the race and if the price is too short I just don't bet. In this day and age with all the tracks to choose from it's not that hard to find a solid play that's overlaid, one must have patience and they will turn up.

I'm going to look at this from another direction.

First, let's assume you are skilled handicapper that is capable of seperating legitimate favorites from false favorites.

If you studied them and broke them out, you would find that the false favorites had a terrible ROI and the legitimate favorites outperformed the take significantly (the net being equal to the take + or minus a little).

If you think about that carefully, it suggests that it's extremely difficult to win money in races where the public has identified the correct favorite.

If legit favorites outperform the take significantly, that means the remainder of the horses in those fields as a group are going to UNDERPERFORM the take signifcantly - making it doubly difficult to find an overlay among them.

That suggests that the best way to play horses is to limit your bets to races where there is a false favorite!

You can usually go with your top choice in those situations because they will be overlays (or at least close) because of the dead money on the favorite and your own skill at eliminating some of the other low ROI dead wood.

All this doesn't mean you can't find overlays betting on your 2nd, 3rd, 4th choice etc.... However, IMHO, even a very skilled handicapper is usually better off sticking with his top choices against false favorites except in certain situations where it's obvious the odds are screwy on someone else.

In my personal wagering, I rarely bet on races where I agree with the odds board about who deserves to be the favorite (some exceptions of course) and bet on my top choice most often in the races with bad favorites.

As to some of the other issues in this thread, I have found that the more comprehensive and skilled I became as a handicapper the more often I agreed with the odds board. Even now, when I disagree, I am WAY more comfortable
that I have an overlay IF I UNDERSTAND the ODDS and know that some of my insights and information are not being built in properly. If I DON 'T understand the ODDS, that usually means there's something "I" don't understand and that I probably "don't" have an overlay.

[1st time lasix]

Classhandicapper has hit the point on the head. I would much rather have the skill and experience to find the actual true vulnerability of the favorite or even the 2nd choice than any other skill in this paramutual game. That is the "dead" money he talks about that is needed by winners to overcome the onerous take out. If the public has it right.....no play. Move on. The "action" of chasing false overlays is what drives most players into the dirt. Nothing could be more fundamental yet for a gambler/recreational player....sitting on your hands takes some real dicipline.

[Scotchman]

Classhandicapper has hit the point on the head. I would much rather have the skill and experience to find the actual true vulnerability of the favorite or even the 2nd choice than any other skill in this paramutual game. That is the "dead" money he talks about that is needed by winners to overcome the onerous take out. If the public has it right.....no play. Move on. The "action" of chasing false overlays is what drives most players into the dirt. Nothing could be more fundamental yet for a gambler/recreational player....sitting on your hands takes some real dicipline.

The other side of this coin is that there are those who search for legitimate favorites, and use them as keys in exotics. This strategy offers a higher hit percentage. But we're talking about the same thing. It is a definite edge to know what is and what isn't a true favorite.

[Tom]

It certainly is! ;)

[classhandicapper]

The other side of this coin is that there are those who search for legitimate favorites, and use them as keys in exotics. This strategy offers a higher hit percentage. But we're talking about the same thing. It is a definite edge to know what is and what isn't a true favorite.

Yes, this approach works well also. Legitimate favorites outperform the take. If you hook them up in exactas, doubles etc... leaving off other very overbet horses it can take you over the top.

[hdcper]

Totally agree with both sides of the coin. Classhandicapper your post was excellent!!!

Well said,

Hdcper

[Bill Olmsted]

Identifying bogus favorites is not easy. You have to be right a high percentage of the time to make it work. To see how hard it really is, try to identify false favorites for at least 100 races and record your results. I think you will be surprised at how good the public really is.

[The Judge]

You have got that right thats the devil of it all. You have the take-out then you have the public the you have to respect and yet whenever possible you must bet agaisnt them. This is the game we choose to play. :D

[Bill Olmsted]

It's a double edged sword when you try to eliminate the favorite (and by that, I mean specifically focusing on ELIMINATING the favorite as opposed to just picking winners). Not only do you have to beat the public by consistently eliminating the favorite successful but you also have to come up with the winner! That's two handicapping chores instead of just one (picking the best horse). I've tried it. Not easy.

[classhandicapper]

Identifying bogus favorites is not easy. You have to be right a high percentage of the time to make it work. To see how hard it really is, try to identify false favorites for at least 100 races and record your results. I think you will be surprised at how good the public really is.

I agree with you. I don't think there are many clear cut false favorites on each racing card. I know I don't find that many in NY and in major stakes around the country. That's why I think that horseplayers that make a lot of bets are often just spinning their wheels. Most of their profits come from a small percentage of total plays.

[hdcper]

I too agree, false favorites are difficult to find, just as finding solid favorites to wager on. Below I first have summarized from my 2007 database, how all favorites do as a group:



Surface: (ALL*) Distance: (All*) (From Index File: C:\JcapperDatabases\GoldJPR2007\pl_Favorites_1.txt )


Data Summary Win Place Show
Mutuel Totals 26883.70 28401.80 28433.70
Bet -32808.00 -32808.00 -32808.00
Gain -5924.30 -4406.20 -4374.30

Wins 5688 9087 11084
Plays 16404 16404 16404
PCT .3467 .5540 .6757

ROI 0.8194 0.8657 0.8667
Avg Mut 4.73 3.13 2.57


By: 21 TheOdds

>=Min <Max Gain Bet Roi Wins Plays Pct
0.00 0.00 0.00 0.00 0.0000 0 0 .0000
0.00 0.50 -126.90 986.00 0.8713 322 493 .6531
0.50 1.00 -1104.40 6048.00 0.8174 1434 3024 .4742
1.00 1.50 -1476.80 8546.00 0.8272 1608 4273 .3763
1.50 2.00 -1572.00 8112.00 0.8062 1219 4056 .3005
2.00 2.50 -938.70 5650.00 0.8339 743 2825 .2630
2.50 3.00 -567.00 2488.00 0.7721 263 1244 .2114
3.00 3.50 -65.60 782.00 0.9161 86 391 .2199
3.50 4.00 -52.70 164.00 0.6787 12 82 .1463
4.00 4.50 -26.00 26.00 0.0000 0 13 .0000
4.50 5.00 7.80 4.00 2.9500 1 2 .5000
5.00 5.50 -2.00 2.00 0.0000 0 1 .0000
5.50 6.00 0.00 0.00 0.0000 0 0 .0000
6.00 6.50 0.00 0.00 0.0000 0 0 .0000
6.50 7.00 0.00 0.00 0.0000 0 0 .0000
7.00 7.50 0.00 0.00 0.0000 0 0 .0000
7.50 8.00 0.00 0.00 0.0000 0 0 .0000
8.00 8.50 0.00 0.00 0.0000 0 0 .0000
8.50 9.00 0.00 0.00 0.0000 0 0 .0000
9.00 9999.00 0.00 0.00 0.0000 0 0 .0000




Now here are the results of all the false favorites I could find this year, roughly only 3 out of every 100 races (not many but they do appear to exist).
Realize they still win just over 25% of the time, but do offer opportunities when going off at 5/2 or less (based on the odd range breakdown):




Dirt (All*) Distance: (All*) (From Index File: C:\JcapperDatabases\GoldJPR2007\pl_Favorites_1.txt )


Data Summary Win Place Show
Mutuel Totals 723.30 795.60 775.50
Bet -1030.00 -1030.00 -1030.00
Gain -306.70 -234.40 -254.50

Wins 130 222 278
Plays 515 515 515
PCT .2524 .4311 .5398

ROI 0.7022 0.7724 0.7529
Avg Mut 5.56 3.58 2.79


By: 21 TheOdds

>=Min <Max Gain Bet Roi Wins Plays Pct
0.00 0.00 0.00 0.00 0.0000 0 0 .0000
0.00 0.50 -2.40 8.00 0.7000 2 4 .5000
0.50 1.00 -24.40 70.00 0.6514 13 35 .3714
1.00 1.50 -78.40 212.00 0.6302 30 106 .2830
1.50 2.00 -82.60 266.00 0.6895 34 133 .2556
2.00 2.50 -91.90 258.00 0.6438 26 129 .2016
2.50 3.00 -12.60 152.00 0.9171 19 76 .2500
3.00 3.50 1.60 48.00 1.0333 6 24 .2500
3.50 4.00 -14.00 14.00 0.0000 0 7 .0000
4.00 4.50 -2.00 2.00 0.0000 0 1 .0000
4.50 5.00 0.00 0.00 0.0000 0 0 .0000
5.00 5.50 0.00 0.00 0.0000 0 0 .0000
5.50 6.00 0.00 0.00 0.0000 0 0 .0000
6.00 6.50 0.00 0.00 0.0000 0 0 .0000
6.50 7.00 0.00 0.00 0.0000 0 0 .0000
7.00 7.50 0.00 0.00 0.0000 0 0 .0000
7.50 8.00 0.00 0.00 0.0000 0 0 .0000
8.00 8.50 0.00 0.00 0.0000 0 0 .0000
8.50 9.00 0.00 0.00 0.0000 0 0 .0000
9.00 9999.00 0.00 0.00 0.0000 0 0 .0000




Finally, he are solid favorites I identified during 2007, against a very small group, but certainly worthwhile when they occurred.




Surface: (ALL*) Distance: (All*) (From Index File: C:\JcapperDatabases\GoldJPR2007\pl_Favorites_1.txt )


Data Summary Win Place Show
Mutuel Totals 419.20 371.50 315.50
Bet -336.00 -336.00 -336.00
Gain 83.20 35.50 -20.50

Wins 98 127 130
Plays 168 168 168
PCT .5833 .7560 .7738

ROI 1.2476 1.1057 0.9390
Avg Mut 4.28 2.93 2.43


By: 21 TheOdds

>=Min <Max Gain Bet Roi Wins Plays Pct
0.00 0.00 0.00 0.00 0.0000 0 0 .0000
0.00 0.50 -0.20 14.00 0.9857 5 7 .7143
0.50 1.00 15.90 98.00 1.1622 33 49 .6735
1.00 1.50 36.90 118.00 1.3127 36 59 .6102
1.50 2.00 19.10 66.00 1.2894 16 33 .4848
2.00 2.50 17.50 34.00 1.5147 8 17 .4706
2.50 3.00 -6.00 6.00 0.0000 0 3 .0000
3.00 3.50 0.00 0.00 0.0000 0 0 .0000
3.50 4.00 0.00 0.00 0.0000 0 0 .0000
4.00 4.50 0.00 0.00 0.0000 0 0 .0000
4.50 5.00 0.00 0.00 0.0000 0 0 .0000
5.00 5.50 0.00 0.00 0.0000 0 0 .0000
5.50 6.00 0.00 0.00 0.0000 0 0 .0000
6.00 6.50 0.00 0.00 0.0000 0 0 .0000
6.50 7.00 0.00 0.00 0.0000 0 0 .0000
7.00 7.50 0.00 0.00 0.0000 0 0 .0000
7.50 8.00 0.00 0.00 0.0000 0 0 .0000
8.00 8.50 0.00 0.00 0.0000 0 0 .0000
8.50 9.00 0.00 0.00 0.0000 0 0 .0000
9.00 9999.00 0.00 0.00 0.0000 0 0 .0000




Just food for thought,

Hdcper

[Robert Fischer]

the public doesn't how to handicapp

[Wizard of Odds]

the public doesn't how to handicapp

Can you get more than 1/3 right betting on one horse for every race?

[singunner]

This topic shouldn't have taken so long to explain. The reality is very simple. Both you AND the mob are accurate. If you look at all your 3/1 shots that people bet into the ground, you'll find they did better than 3/1. If you look at all your 3/1 shots that are getting little action, you're going to find they come in less often.

The fact of the matter is that the public isn't shooting completely in the dark. BUT, if you're losing money in one direction, I'd wager you're losing money in the other too. There could be a sweet spot (imagine two curves on a graph where they have two intersect points) in which you should be placing your bets, or there could be only one intersect point, so you should bet no matter what the odds.

Here's a simple real-world example using my program (which is more basic than anything any right-minded horseplayer uses):

If odds don't matter, my predicted winner comes in 27.1% of the time.
If odds are lower than I predict, that goes to 31.8%
If odds are higher than I predict, it drops to 11.5%

And here's what would be most pertinent to your query:
Odds are 0% to 10%> what I predict, 12.3% ROI 96.6%
Odds are 0% to 20%> what I predict, 11.5% ROI 97.7%
Odds are 0% to 50%> what I predict, 11.5% ROI 97.4%

Yes, I win slightly more frequently when I have an odds cap, but it actually lowers my ROI by 1%. This is run across tens of thousands of races, so I'm pretty confident in the results. I obviously have a sytem that doesn't have a sweet spot (two intersections between the curves).

[MichaelNunamaker]

Hi Bill,

You wrote "It's a double edged sword when you try to eliminate the favorite (and by that, I mean specifically focusing on ELIMINATING the favorite as opposed to just picking winners). Not only do you have to beat the public by consistently eliminating the favorite successful but you also have to come up with the winner! That's two handicapping chores instead of just one (picking the best horse)."

While I certainly agree that it is hard to eliminate a favorite, it is worth noting that if one can reliably eliminate the favorite, no other handicapping is needed. If a favorite really does have no (or a very low) probability of winning, then every other horse in the race will be an overlay. Simply bet them all.

That said, I don't know anyone who can point at any non-trivial sized group of favorites that win at a low single digit percentage. It's a very very tough problem and one that potentially has no solution.

Mike Nunamaker

[Overlay]

While I certainly agree that it is hard to eliminate a favorite, it is worth noting that if one can reliably eliminate the favorite, no other handicapping is needed. If a favorite really does have no (or a very low) probability of winning, then every other horse in the race will be an overlay. Simply bet them all.

That said, I don't know anyone who can point at any non-trivial sized group of favorites that win at a low single digit percentage. It's a very very tough problem and one that potentially has no solution.

Even if you can't totally eliminate the favorite from consideration, any degree of overbetting on it will raise the odds of the other horses in the race. Also, a horse can be a legitimate favorite, and still be overbet in comparison to its actual chances. An odds line can help in detecting both those types of situations, as well as determining which other horses in the field are benefitting the most with regard to the relationship of their odds to their true probability of winning, as a result of the excessive support of the favorite.

[sjk]

I don't know why one would want to look only at favorites or only at a specific odds range.

At every odds level I find 9-10% of the horses sufficiently overlaid to be playable. Rather than sit around waiting for horses at low odds to come along I think it is maximizes my results to play them all.

Of course I would tend to bet more on those that I think have the greater probability of winning and those that I think offer the greater overlay factor (getting back to the original subject).

I expect to be wrong more often when the overlay factor is high but the extra payoff more than compensates.

[Robert Fischer]

Can you get more than 1/3 right betting on one horse for every race?

like an arbitrary win% contest with the public?
I wouldn't last a day trying to handicap every race. They would overwhelm me with sheer volume.
I waited what seemed like 2 months for commentator to run back and I didn't have a win bet on High Finance but my horses came in the exacta and trifecta.

[Kelso]

That said, I don't know anyone who can point at any non-trivial sized group of favorites that win at a low single digit percentage.


Just thinkin' out my butt here ... might low-priced claiming races ... or low-priced claimers at cheaper tracks ... or low priced claimers with no major class-drop runners ... do the trick? "Less likely to run to form" sort of thing?

[classhandicapper]

I don't know why one would want to look only at favorites or only at a specific odds range.

At every odds level I find 9-10% of the horses sufficiently overlaid to be playable.


It's not much that you should focus just on favorites. It's that legitimate favorites as a group outperform the take. By definition they have a large percentage of the pool bet on them. If a large part of the betting pool will outperform the take, the remainder MUST underform the take.

If I were to ask you whether you'd like to play the horses at a track with a 20+% take, I'm sure you'd say no. But that's exactly what you are often doing when you start fishing for overlays among long shots when the favorite is legit. That's a pretty tough game.

I am only pointing out a way of thinking about this game that I am pretty certain will help people raise their ROI and profits. They need to skip more races.

(of course this is only relevant if you have the skill to identify bad favorites)

[sjk]

I love to play Pha where the takeout is higher.

Higher takeout may mean that fewer bets make the cut but the ones that do are still worthwhile.

[NoDayJob]

Since my software can pick up to 3 contenders, if only one horse has morning line odds of 2-1 or less or more than 20-1, it is eliminated, and the remaining two selections are played. Otherwise, there's no limit on M.L. odds if only one or two horses are contenders. I never look at the "Ironman" odds either.

[Kelso]

It's not much that you should focus just on favorites. It's that legitimate favorites as a group outperform the take. By definition they have a large percentage of the pool bet on them.


Is there a commonly accepted way of identifying this "group," such that meaningful stats can be drawn from/about it?

I apologize if I've missed a definition to the contrary, but favorites seem to become "legitimate" only after they've run well. What is the definition of "legitimate favorite?"

Thank you.

[betchatoo]

A legitimate favorite is one that wins in a proportion equal to, or better than, its odds. Unfortunately, there is no one way to determine if a favorite is legitimate. It depends on your individual handicapping. I use my software which is very good at picking out poor favorites. I also have a couple of UDMs (spot plays) that tell me if a favorite is legit.

[Kelso]

A legitimate favorite is one that wins in a proportion equal to, or better than, its odds. Unfortunately, there is no one way to determine if a favorite is legitimate. It depends on your individual handicapping.


That 'splains it. Thank you.

[Bill Olmsted]

Hi Bill,

While I certainly agree that it is hard to eliminate a favorite, it is worth noting that if one can reliably eliminate the favorite, no other handicapping is needed. If a favorite really does have no (or a very low) probability of winning, then every other horse in the race will be an overlay. Simply bet them all.


Mike Nunamaker

Hi Mike,

Long time no see. I've found that if you can eliminate the favorite then simply betting the crowd's next choice (the 2nd favorite) is the best (ROI-wise) strategy. I tried this for six months and had decent results. Of course, the lower the favorite eliminated the better payoff on the 2nd favorite (or all remaing horses for that matter).

Of course, you could "bet them all" (I've tried) but remember this: If your false favorite happens to win--and the field has 10 horses, that means you could be out nine betting units! That is significant loss and a huge nut to overcome. Bottom line: betting the field only works if you can eliminate the favorite a high percentage of the time. The best course of action is to bet the crowd's second betting choice.

[classhandicapper]

Is there a commonly accepted way of identifying this "group," such that meaningful stats can be drawn from/about it?

I apologize if I've missed a definition to the contrary, but favorites seem to become "legitimate" only after they've run well. What is the definition of "legitimate favorite?"

Thank you.

I wouldn't classify a favorite as legitimate if he wins or runs well (and vice versa).

If you have the skill to identify a group of favorites that produces an ROI that's much worse than the track take, you have some very valuable information to work with (and vice versa).

[Bill Olmsted]

A favorite is only "legitimate if it produces a higher ROI than favorites as a whole.

[Tom]

I don't think that is right - by that definition, wouldn't all favoites that win but pays less than all favs as a whole be not legit? The roi is based on one payoff.

[Bill Olmsted]

I don't think that is right - by that definition, wouldn't all favoites that win but pays less than all favs as a whole be not legit? The roi is based on one payoff.

You are right. It is impossible to tell if any one single favorite--or any odds horse for that matter--is legitimate (e.g."overlaid"). I stand corrected.

[Cratos]

I don't know why one would want to look only at favorites or only at a specific odds range.

At every odds level I find 9-10% of the horses sufficiently overlaid to be playable. Rather than sit around waiting for horses at low odds to come along I think it is maximizes my results to play them all.

Of course I would tend to bet more on those that I think have the greater probability of winning and those that I think offer the greater overlay factor (getting back to the original subject).

I expect to be wrong more often when the overlay factor is high but the extra payoff more than compensates.

I believe that you are confusing “favorite” with “winner”. Every favorite is legitimate because it is a before-the-race opinion established by the wagering public. Every race has a winner(s) and every race has a favorite(s), but every favorite is not the winner of the race. Obviously after the race and a longshot has won, it can be said that the longshot should have been the favorite, but coulda, shoulda, and woulda don’t win races.

[Overlay]

You are right. It is impossible to tell if any one single favorite--or any odds horse for that matter--is legitimate (e.g."overlaid"). I stand corrected.

I'm not sure I agree with that statement. I would definitely agree that, before a race is run, it's impossible to say with certainty that any particular horse in it (whether it's the favorite or not) has a 100% guarantee of winning. But there are reliable ways to judge whether a horse has a better chance of winning a race than its odds indicate.

[cj]

You can certainly find conditions where the favorite is likely false. I'll give one example. Using my numbers, if the horse's last race on the same surface as today doesn't put it in the top 3 in the race, that is a false, or at the least very vulnerable, favorite.


Sts Wins WinPct AvgPay ROI
7053 1786 25.32% $5.38 68.17%


So, not only are these races to not bet the favorite, they are races to attack. There are many other ways to find vulnerable favorites.

Here is one more, favorites in two turn races breaking from the 8 post or wider:


Sts Winners WinPct AvgPay ROI
1181 320 27.10% $5.35 72.42%

[ryesteve]

But there are reliable ways to judge whether a horse has a better chance of winning a race than its odds indicate.
Perhaps I'm wrong, but I believe the point was that even after the outcome is known, a single race can't confirm or refute that opinion.

[classhandicapper]

You can certainly find conditions where the favorite is likely false. I'll give one example. Using my numbers, if the horse's last race on the same surface as today doesn't put it in the top 3 in the race, that is a false, or at the least very vulnerable, favorite.


Sts Wins WinPct AvgPay ROI
7053 1786 25.32% $5.38 68.17%


So, not only are these races to not bet the favorite, they are races to attack. There are many other ways to find vulnerable favorites.

Here is one more, favorites in two turn races breaking from the 8 post or wider:


Sts Winners WinPct AvgPay ROI
1181 320 27.10% $5.35 72.42%


Great stuff and exactly what I am taking about. :ThmbUp:

Now let's say you had 10 conditions like that and you excluded all those vulnerable/false favorites. The remaining favorites would almost certainly outperform the take even if they still showed a small loss.

Playing against those more legit favorites would be putting yourself into dangerous (or at least more difficult) territory because the remaining horses in those fields would lose more than the take as a group making it much harder to find an overlay. Kind of like playing horses with a higher take.

[Kelso]

Using my numbers, if the horse's last race on the same surface as today doesn't put it in the top 3 in the race,

Here is one more, favorites in two turn races breaking from the 8 post or wider:


CJ,
Thank you for sharing these two false-favorite indicators. If I'm not pushing my luck by asking .....

1) Does the "last race" situation also require same distance, or is surface the only condition involved? If distance also, would it be exactly the same distance, or +/- a furlong or so?

2) Does surface matter to the "two turn" situation?

Thank you, again.

[Overlay]

In theory, I agree there can't be "too much of an overlay" but over the course of thousands of events Benter and others have detremined that the "best" odds-line will be about 55% your line (assuming you make a good line) and 45% the public's line.

If you had an odds line that was 100% accurate, every horse that went off at odds that were any amount higher than the odds in your line would be a true overlay, without having to demand any type of "premium" on your odds in order to compensate for unknown or uncertain conditions. But under the conditions stated in the quote above, where the "true" odds would be a composite of 55% of your line and 45% of the public's line, what degree or percentage of overlay would you have to demand on the odds in your line before the horse would be a true overlay?

[sjk]

Overlay,

I have no taste for revisiting the math but I think my line comes pretty close to the description that 55% of my line plus 45% of the public line has the best (less than perfect) accuracy of linear combinations of the two.

I use an 80% overlay requirement which I think is mathematically equivalent to using a much thinner percentage (maybe 20%) for the above composite line.

At the margin there is a trade-off between total dollars won (reduce the requirement for this) and ROI (increase it). If you try to maximize dollars won by accepting a lower ROI you will have to live with longer losing streaks and I'm not sure that is a good trade-off for a very modest increase in dollars.