How to play top 3 selections...

[AutumnLotus]

Hey Everyone,

I'm certain this topic has more than likely been brought up throughout the years however I was unable to come up with any threads via the search option I think probably more than likely due to not knowing how to word it properly or use proper keywords, so please accept my apologies up front if this has been a repetitive topic.

My question is based around an observation I have made many years ago and it is certainly an obvious one;

For example, I'm sure you've all noticed let's say looking at pic services such as Racingdudes or Equinedge as well as pics that are handicapped and displayed at each track from the tracks resident handicapper(s), there is always three selections. No that's not to say of course that the handicappers displaying those three will not give you any direction on which to play or rather how to play i.e., they Their first selection out of the three for the win or two say some combination of an exotic wager as well as when Etcetera we've got my point.

In my opinion and observation this is how it goes across the board, even when I do my own handicapping I almost always end up with three selections. It is even common practice here on the board when people place their opinions it is almost always 3 horses.

I guess my question is, how do you play your selections when you have three because we know it is certainly is not going to be a trifecta play. This is something that I have been struggling with personally as I have repeatedly saw that in most cases one of the three Selections that I come up with will have one of them come in for the win and if not certainly 1 if not two of them will be in the money. I guess I'm trying to figure out what is the best way to go about playing these selections. I'm aware of course that that is a very wide open question obviously based on how one values each of the three selections. But I'm asking it more so to have this one go about it if those three selections all seem viable for the win?

[Poindexter]

Every race is different. To begin with only bother with races of at least 7 horses.
Which likely means you need to wait to initial scratches are posted to determine where you are betting. Now after handicapping a race rate your top choice by strength. Standout, very strong, strong, solid, clear, weak.From a fair odds perspective on a 7 horse field that might mean
3/5, 6/5. 8/5, 2-1, 5/2, 3-1 to 7/2.

On a 10 horse field and up that might mean

1-1, 8/5, 2-1, 5/2, 3-1, 4-1

Probably around 10 years ago I posted a formula I created about 30 years ago to get a ball park fair odds line for your top 3 contenders. Maybe search no line bet chart and see if it comes up. It unfortunately is not in my current computer. If you find it and use it for every race you handicap, you can save yourself the trouble of having to line the entire field. That at least will provide you with a shortcut way to determine if any of your top 3 (I usually go to 4 selections) our good value plays based off of your opinion (perceived value). So at least at that point you are able to identify good win bets. You should want at a 20 percent premium over fair value). We had a fairly recent discussion on place betting and I was outvoted by most of the sharps in this room so stick with win betting until your data convinces you otherwise.

By the way it is unlikely that you will find value on standouts. In my opinion these horses are much better plays as keys in both verticals and horizontals. If you go through the process of using my no line bet chart (I will find it and post it when I can), set up a spreadsheet on every horse you have handicapped.
Your line, odds at load, final odds, distance, surface, class. If you want to go further you can add jockey pctg, trainer pctg, reason you liked the horse, morning line, with or against bias, physicality and anything else you can think of. Obviously every bit of data creates more work so you may want to limit the extra data collection to just your bets.

Now the next question is exotics. I will start with exactas. The problem of course is whales. In 1995 you could look at the tv monitor at post time and if the 2-7 exacta was paying $21 chances are you were getting close to that amount. Now, who knows. To determine fair exacta price based off of the fair odds you have created take the odds of the top horse multiply by the odds of the second horse then add the odds of the first horse. Multiply by 1 for $2 exactas and 2 for $2 exactas. So if you have the 2 at 6/5 and the 6 at 6-1 the fair exacta payoff is $8.40 foot a dollar. Reversed would be $13.20. So there may be times that there is no win value but perhaps there might be exacta value. With exactas you may want to require 30 percent above fair value.

Now you forget the last paragraph and just focus on win betting with exacta covers. I pretty much do. You mentioned you typically have 3 contenders. You pick the race 3-4-6. 3 is 8/5, 4 is 3-1, 6 is 10-1. So you opt to bet $20 to win on the 6. You can also play a $8 exacta 3-6, a $6 exacta 4-6, and maybe a 50 percent reverse (6-3, 6-4). Of course this means you have a separate exacta bankroll, but you are protecting win bets on the bulk of your play so the hedge nature limits your bankroll exposure. If you chose to bet win place I would treat exactas as a completely separate bankroll.

I am going to end it here. I would get well versed at the above before going deeper. If you want to get your feet wet, most tracks have a 50 cent minimum. So play a 50 cent box of your top 3 horses a $1 key with your top horse and a $1 cold deck. Basically it will cost you $6 a race and see how you do.

By the way there are 2 reasons why touts typically do not provide a lot of direction, one it is very easy to make yourself look good if you post 3 horses a race with no direction. They certainly do not want to lock themselves into an odds line because many of their winners will be below fair value. Also in fairness to them it is not their job to think for you. Everybody plays differently. So 10 people can use the same tout 2 have a good day, 3 do okay, 3 have a rough day, 2 get crushed. The tout himself can also fit into any of these categories.

[Poindexter]

Unable to find the post I made. But I gave you a blueprint for estimating the chances of your top choice. You can probably use around 33 percent to 35 percent as the probability of your next 2 choice winning. So if my top choice is 6/5 typically I would have a 2nd choice at 4-1 and a 3rd choice of 6-1 which totals 34 percent or maybe 3-1 and 8-1 which totals 36 percent. If I make my top choice 2-1 then my second choice might be 4-1 and my third choice might be 6-1. Obviously with standouts you might have a 6-1 2nd choice and a 8-1 3rd choice thus they would account for only 25 percent. But remember if your top choice is 3/5 you are assigning Him 62.5 percent and you have to leave room for the other horses.

So bottom line pick an odds line that feels right to you on all 3 horses, add up the probability you have assigned the 3 horses, make sure it is reasonable. On competitive races 60 to 65 percent is a good number. Something like 3-1, 4-1, 5-1(about 61.7 percent). More obvious race might be 2-1, 3-1, 5-1 (which is 74.7 percent which is not easily attained so make sure you are not underestimating the rest of the field. Over time you will figure it out. Meanwhile your database will provide you the tools to determine how well you are doing and ultimately how much margin you need to look for to maximize performance (which initially or ultimately may mean losing less and not winning).

By the way you are not limited to your top 3 choices. I am just addressing your
question.

[Saratoga]

Originally Posted by AutumnLotus

Hey Everyone,

I'm certain this topic has more than likely been brought up throughout the years however I was unable to come up with any threads via the search option I think probably more than likely due to not knowing how to word it properly or use proper keywords, so please accept my apologies up front if this has been a repetitive topic.

My question is based around an observation I have made many years ago and it is certainly an obvious one;

For example, I'm sure you've all noticed let's say looking at pic services such as Racingdudes or Equinedge as well as pics that are handicapped and displayed at each track from the tracks resident handicapper(s), there is always three selections. No that's not to say of course that the handicappers displaying those three will not give you any direction on which to play or rather how to play i.e., they Their first selection out of the three for the win or two say some combination of an exotic wager as well as when Etcetera we've got my point.

In my opinion and observation this is how it goes across the board, even when I do my own handicapping I almost always end up with three selections. It is even common practice here on the board when people place their opinions it is almost always 3 horses.

I guess my question is, how do you play your selections when you have three because we know it is certainly is not going to be a trifecta play. This is something that I have been struggling with personally as I have repeatedly saw that in most cases one of the three Selections that I come up with will have one of them come in for the win and if not certainly 1 if not two of them will be in the money. I guess I'm trying to figure out what is the best way to go about playing these selections. I'm aware of course that that is a very wide open question obviously based on how one values each of the three selections. But I'm asking it more so to have this one go about it if those three selections all seem viable for the win?

Thank you so much for stating what I've been saying all along....

There's guys that post picks here with 4 ,5, sometimes 6 selections without a strategy on how to play them....
I'm OK with that but what bothers me is when they provide results like hitting and exacta , triple , super and say they had a GREAT DAY!!!
Figuring out what you have to lay out to cover all these bets doesn't even come close to profit on what they posted they won....
There's one kid here selling picks that does exactly that and I say to myself get a real job before your too late ......
When I make picks I tell you exactly how I bet , so there's no doubt if I won or lost..
There's another guy thats honest too..5THSTREETPICKS...

[DanBoals]

Hey Autumn Lotus,


There is the idea put forward by Danny Holmes on how to bet your top three choices:


the thread is:


http://www.paceadvantage.com/forum/s...ad.php?t=50961


and I think his book is here, but is also linked in the thread:


http://www.paceadvantage.com/forum/a...7&d=1222428231


My advice would be to keep records on your first, second and third choices as well as their odds and then see empirically what shows a profit.


For example, if you bet the two longest odds horses of your three choices, does that show a profit? If you have the results recorded, you can answer that.


Best of luck,

[Dave S]

Just in terms of win betting an older book by Henry Kuck called "Winner's File"
made a very succinct recommendation which I liked. I am paraphrasing here but he strongly recommended sticking with one of your top two selections unless you were getting exceptional odds on your third choice: e.g. 10:1 or higher. Of course if both your top selection's odds are high enough then bet both.

[MJC922]

One thing I've noticed over the years is that anything any model over a large sample that makes money, and even consider a huge backfit for purposes of this discussion, does not make money with the model's third choices. Take it FWIW.

[Dave S]

Originally Posted by MJC922

One thing I've noticed over the years is that anything any model over a large sample that makes money, and even consider a huge backfit for purposes of this discussion, does not make money with the model's third choices. Take it FWIW.

Agreed. But I would think that higher odds 3rd ranked choices should have much better hit rates for place and show than for win only if the method for picking them has any validity.

[MJC922]

Originally Posted by Dave S

Agreed. But I would think that higher odds 3rd ranked choices should have much better hit rates for place and show than for win only if the method for picking them has any validity.

It's a weird phenomena imo but seen consistently enough that it's worth pondering. The thing of it is, the more predictive the model the worse third rank does, so much so that third becomes an increasingly negative trait. It's kind of like when you think of what Quirin did with one of his spot play 'systems' i.e. ranked one two in speed figures and then ranked one two in earnings figures, so the value is in being ranked top two in some model, maybe an entirely different type of model altogether, pace, physicality, connections, whatever, but just from a high level it seems as if something better than third is usually where the value seems to be concentrated. I see what you're getting at with regard to price, I mean such and such is ranked third on class and is 50-1, probably at least some value mixed in there even if third isn't really all that 'positive' in general, it's still probably good enough to win 5% of the time at that number and underneath in exotics might be worth looking at.

[AutumnLotus]

Originally Posted by Poindexter

Unable to find the post I made. But I gave you a blueprint for estimating the chances of your top choice. You can probably use around 33 percent to 35 percent as the probability of your next 2 choice winning. So if my top choice is 6/5 typically I would have a 2nd choice at 4-1 and a 3rd choice of 6-1 which totals 34 percent or maybe 3-1 and 8-1 which totals 36 percent. If I make my top choice 2-1 then my second choice might be 4-1 and my third choice might be 6-1. Obviously with standouts you might have a 6-1 2nd choice and a 8-1 3rd choice thus they would account for only 25 percent. But remember if your top choice is 3/5 you are assigning Him 62.5 percent and you have to leave room for the other horses.

So bottom line pick an odds line that feels right to you on all 3 horses, add up the probability you have assigned the 3 horses, make sure it is reasonable. On competitive races 60 to 65 percent is a good number. Something like 3-1, 4-1, 5-1(about 61.7 percent). More obvious race might be 2-1, 3-1, 5-1 (which is 74.7 percent which is not easily attained so make sure you are not underestimating the rest of the field. Over time you will figure it out. Meanwhile your database will provide you the tools to determine how well you are doing and ultimately how much margin you need to look for to maximize performance (which initially or ultimately may mean losing less and not winning).

By the way you are not limited to your top 3 choices. I am just addressing your
question.

Thank you very much for taking the time to write out such a detailed response, truly I appreciate it and your suggestions will not go unnoticed.

Although I do want to say this, your suggestion of essentially assigning a percentage of probability between the three selections, essentially ranking them is a fine suggestion however one of the troubles that I have is doing exactly that; for the most part when I handicap a race and settle upon my three selections I honestly for the life of me cannot discern, or rather, I cannot see them as anything other than all three having the same chance of winning. I know that may come off as a bit odd or silly to say, but for the most part every time I attempt to assign a percentage of probability to separate them into rank and order I end up getting much too analytical, to the point where my line of confidence begins to muddy my decision making.
The problem I have is that I'll look at for example the first horse out of the 3 and I'll say something like this horse has the best speed figure by far, then I'll look at the second horse out of the three and I'll say something like but this worse has the best speed at distance today as well as the highest prime power number, then I'll look at the third horse and it'll be something like this horse today is being ridden by a high percentage winning jockey with a positive jockey and trainer combo.

I'm just throwing out super quick examples I'm sure you understand that I'm simply trying to say that I completely agree with your suggestion but I literally have a very hard time making solid decisions and assigning one has more value between the three in order to rank them appropriately.

Another very important point that I also want to make here is that At the end of the day I will always go back and look At the horses that have won and then I will look Over its past performance as well as my data to see if there is anything that stands out that I should have caught or that I know to look for next time and what I discover is that between the three selections the ones that I would think should have came in doesn't and the horse that did come in has Nothing of value that would suggest this horse over the other. I hope I am making sense. What I'm saying is when I go back and analyze what it is that made one horse out of my three come in for the win over the other two it seems to me that I would never have chose that horse to be the winner or to be the better of the other two.

[Dave Schwartz]

What an excellent post.
(I pushed your quote down simply because there was so much.)

From what you've posted it seems that you handicap, then you do a review of your day, trying to discern what you:

* Did well
* Could have improved
* Did Poorly

The fact that you are willing to put forth the effort to look objectively at your day says a lot about your desire to improve.

Imagine you are watching a stock market analyst on TV after the markets have closed.

They describe what happened on the day (i.e. "The Dow went up today because...").

We really have no idea whether the analyst's reason for the movement was right or wrong.

Was anything really learned by what the analyst said?

Unlike you, he does not post his selection BEFORE the day's events. If he did, you could track his selections and determine if his opinions are even worth listening to.

If he did post his selections ahead of time, you could accumulate some data and analyze that.

However, without an accumulation of data to study, learning is always very difficult.

While this is an over-simplification, it is what must be done.

If you really want to improve your game,
you need the answers to some very important questions. Here are some general examples.)

• "How well do your assessments perform over (say) a 100-race sample?"
• "How accurate are the base components that lead you to your decisions?"
• "Are those base components really important?"

There really are no shortcuts to this that I can see.

Hope this helps.


Dave
___

Originally Posted by AutumnLotus

Thank you very much for taking the time to write out such a detailed response, truly I appreciate it and your suggestions will not go unnoticed.

Although I do want to say this, your suggestion of essentially assigning a percentage of probability between the three selections, essentially ranking them is a fine suggestion however one of the troubles that I have is doing exactly that; for the most part when I handicap a race and settle upon my three selections I honestly for the life of me cannot discern, or rather, I cannot see them as anything other than all three having the same chance of winning. I know that may come off as a bit odd or silly to say, but for the most part every time I attempt to assign a percentage of probability to separate them into rank and order I end up getting much too analytical, to the point where my line of confidence begins to muddy my decision making.
The problem I have is that I'll look at for example the first horse out of the 3 and I'll say something like this horse has the best speed figure by far, then I'll look at the second horse out of the three and I'll say something like but this worse has the best speed at distance today as well as the highest prime power number, then I'll look at the third horse and it'll be something like this horse today is being ridden by a high percentage winning jockey with a positive jockey and trainer combo.

I'm just throwing out super quick examples I'm sure you understand that I'm simply trying to say that I completely agree with your suggestion but I literally have a very hard time making solid decisions and assigning one has more value between the three in order to rank them appropriately.

Another very important point that I also want to make here is that At the end of the day I will always go back and look At the horses that have won and then I will look Over its past performance as well as my data to see if there is anything that stands out that I should have caught or that I know to look for next time and what I discover is that between the three selections the ones that I would think should have came in doesn't and the horse that did come in has Nothing of value that would suggest this horse over the other. I hope I am making sense. What I'm saying is when I go back and analyze what it is that made one horse out of my three come in for the win over the other two it seems to me that I would never have chose that horse to be the winner or to be the better of the other two.

[MJC922]

It's certainly good to review what came in each day that you handicap and try to discern what you may have done better. Really though you want to eventually reach a point where you actually know what you did was the right thing even when it loses. Otherwise you're going to be one of those people who has pretty much a new tweaked system/method with every new card that comes along based upon what 'worked' in those 8 or 9 races they ran yesterday. Everyone really should study the predictive value of the inputs that they use, so if you use prime power, know how often it wins, if you use Beyer speed figures have a very good idea of how often they hold up on all surfaces. Know how often these types of races are won wire to wire. Know when the rail is not a good post position for this type of horse etc. How often is this type of trip a winning one? We can go on and on but if you're going to be any good at separating horses these things need to be 'known' to you even if they're typically little details, little things are sometimes decisive and will go right under the radar both before and after the races.

[Poindexter]

Originally Posted by AutumnLotus

Thank you very much for taking the time to write out such a detailed response, truly I appreciate it and your suggestions will not go unnoticed.

Although I do want to say this, your suggestion of essentially assigning a percentage of probability between the three selections, essentially ranking them is a fine suggestion however one of the troubles that I have is doing exactly that; for the most part when I handicap a race and settle upon my three selections I honestly for the life of me cannot discern, or rather, I cannot see them as anything other than all three having the same chance of winning. I know that may come off as a bit odd or silly to say, but for the most part every time I attempt to assign a percentage of probability to separate them into rank and order I end up getting much too analytical, to the point where my line of confidence begins to muddy my decision making.
The problem I have is that I'll look at for example the first horse out of the 3 and I'll say something like this horse has the best speed figure by far, then I'll look at the second horse out of the three and I'll say something like but this worse has the best speed at distance today as well as the highest prime power number, then I'll look at the third horse and it'll be something like this horse today is being ridden by a high percentage winning jockey with a positive jockey and trainer combo.

I'm just throwing out super quick examples I'm sure you understand that I'm simply trying to say that I completely agree with your suggestion but I literally have a very hard time making solid decisions and assigning one has more value between the three in order to rank them appropriately.

Another very important point that I also want to make here is that At the end of the day I will always go back and look At the horses that have won and then I will look Over its past performance as well as my data to see if there is anything that stands out that I should have caught or that I know to look for next time and what I discover is that between the three selections the ones that I would think should have came in doesn't and the horse that did come in has Nothing of value that would suggest this horse over the other. I hope I am making sense. What I'm saying is when I go back and analyze what it is that made one horse out of my three come in for the win over the other two it seems to me that I would never have chose that horse to be the winner or to be the better of the other two.

First off just because the public typically is able to latch on to a top choice in most races and they perform remarkably well does not mean that you or I as handicappers are capable of doing the same thing. Your job is to learn what works for you. You may handicap 20 races and have a strong, discernible opinion in just 3 of them. That is fine. I am going to suggest you focus on the races you have a good opinion on and write up a thorough analysis of the top contenders. At that point you can refer back to that analysis when you do your post race analysis. There are a lot of races that I cannot find much or even any separation between my top contenders.

You mentioned the prime power rating. Since you seem far more comfortable with 3 top contenders, use that as your ranking. Use the type of database analysis I mentioned above and see what your roi is based on the prime power ratings. See if you can find anything that indicates over a fairly large sample that you can reach a roi loss of under 10 percent. From there you can try to build. If everything over the course of time points to a 15 percent to 25 percent roi loss, then you are not going anywhere.

One last note. Track your exacta roi on the 6 possibilities (1-2, 1-3, 2-1…..). At this point I would say you are at the data accumulation point. You are trying to learn if you have or can develop the tools to approach profitability. Keep your betting very small until you get to the point that you are confident that what you are doing is working.

[thaskalos]

If it wasn't for the pesky problem of separating contenders, then this wouldn't be the infuriating hobby/obsession that it is. And there would be less second-guessing...and fewer sleepless nights.

[JohnGalt1]

When I have a race where I have three top choices, I usually bet the two with the highest odds to win and play the third in the exacta, if the odds are high enough.

Example-- 9/5 3/1 5/1

Even if I thought the 9/5 was the most likely winner, betting the longer shots to win is more profitable.