Finding Contenders

[Myhorse1_X]

I always have taken the attitude that the first step in Handicapping is to eliminate the non-contenders, then handicap the remaining horses to determine the winner.

I have a little system that I use with the Beyer Numbers to come up with the contenders.
I use the best 2 of the last 3 beyer numbers add them together and take the top 5 numbers for my contenders.

Does anyone else have a similar method?, I would like to hear from them.

MYHorse1

[kitts]

Gordon Pine has mechanical contender selection method that he explains on his website (I'm pretty sure it uses Beyers):
http://www.netcapper.com

[Tom]

Here's a couple I have used in the past:

Add the Beyers for the last two good races and use the top 4-5 as contenders.

Use the best of last two Beyers, take the top three.
Ignore the last two races, take the best of the remaining Beyers, use the top three. You now have 3-6 contenders. (This was from Scott Finley, given to the Sartinistas in Baltimore, 1993).

Here is one I really like: Find the best last-race Beyer, similar distance, same surface, within 30-days. Anything within 10 points in one of its last two races is a contender.

[Aussieplayer]

Myhorse1: That's the same method that Gordon Pine mentions. (He averages, but adding has same affect).

Tom: Where you say, "Ignore the last two races, take the best of the remaining Beyers, use the top three. You now have 3-6 contenders. " Do you mean the best remaining from anywhere in the pp's? Or do you mean using the third last race?

Tom (again): In the last method you mention, how many lengths (roughly) is 10 Beyers? From memory, about 5?
I think using "within10 points in last 2 races" would make most of the field a contender?

Cheers
AP

[so.cal.fan]

I like to use the best Beyer at the track the horse is running at.
Of course, you can't go back too far, a long memory is not a good thing, in my opinion.
Interestingly, you come up with some prices.
I have done no research on this, so it is only a recent observation, perhaps a mirage.
Anyone have stats?

[GR1@HTR]

A simple and in some cases more effective than most other factors at certain tracks is the top4 mlo. No joke...

[Aussieplayer]

GR1, By doing that you are really limiting yourself to the public's opinion. You are (IMO) going to get biased results at least in a general sense, due to not covering the odds spectrum.

Cheers
AP

[ranchwest]

I remember a long time ago a friend of mine and I would get the contenders by looking at the 4 leading money earners. We called them BMW's, big money winners. Back then, they won about 70% or so of the races.

[dav4463]

I also use Beyers as part of my contender selection. I add the 2 best of the horse's last 3 Beyers plus the horse's best Beyer in his races 4-10 back. Then I average them together. Next I eliminate any horse ranking below 7th. Of these 7 horses, I then eliminate a horse if his average is 16 or more below the best horse in a sprint or 21 or more below the best in a route. Next I eliminate horses with absolutely no early speed unless the race really sets up for a closer. Then I eliminate horses that have a negative or suspicious class move. Finally I eliminate any horse who has a trainer and/or jockey with at least 20 races and a 4% or lower win percentage. Also if both the trainer and jockey or below 9% wins, but above 4%, I eliminate the horse unless his Beyer average ranks in the top 4. I hope this helps. Give it a try and see how it works. You may also eliminate a horse that is an absolutely poor fit at today's distance or similar distance or surface.

[GR1@HTR]

MLO (morning line odds) is set by the a track employee. I think you misunderstood what I mentioned. Not the top 4 lowest odds horses at post time. If you run numbers through a database, certain mlo setters (NYRA for example) are very good.

Lets look at the last 940 races at AQU

MLO Rank
1.........35%
Top 2...57%
Top3...74%
Top4...84%

last 542 at CD

1.......29%
Top2..54%
Top3..70%
Top4..80%

Last 887 at FG
1...........33%
Top2.....53%
Top3.....69%
Top4....83%


For the record, I don't suggest using this method, but it is about as simple as it gets for good results at certain tracks.

[Dick Schmidt]

I posted this on another thread, but it fits better here. For what it's worth:


I think choosing contenders at all is a mistake. I stopped doing it about 1990 and haven't looked back since. I put them all in, and then let my handicapping method rate them. I will bet some of the top few, using a structured, pre-defined approach and the rest eliminate themselves. If you can't trust your method to rank the horses, how can you trust it to bet with?

Dick

[cj]

I agree with Dick, I rate EVERY horse in the field regardless of how they look at first glance.

CJ

[Aussieplayer]

I think choosing contenders at all is a mistake.

Thanks very much Dick, just when I was about to ask your opinion on something!! LOL

I was about to say this might be a good thread to pose a question I've been meaning to. Was especially interesting in Dave's & your opinions.

Doing a bit of research, I came up with a "group of horses" (see, I'm trying to avoid the contender word now, lol) that produced the following, across all races:

Group 1 - The BIG Group:
average 6.4 horses per race (52% of field),
64% of winners (I know, not very high)
IV of 1.23
ROI of 12%

IV of non-contenders (oops!) is 0.75

Group 2 - The REFINED Group: (the contenders really!)
Average 3.5 horses per race (29% of field)
42% of winners
IV of 1.46
ROI of 16%

IV of non-contenders is 0.81

Now, I haven't ranked them yet, or broken down into class/distance etc. They are simply all the races I would normally be interested in betting (ie. Saturday races - some of us work!)

At first I thought, "cool" - then I compared the 42% of winners to your 54% and wasn't so impressed. Though 3.5 horses per race probably converts to 2.9 for you - guessing at average fields of 10 for you.
At least the ROI side is looking okay I guess.

Anyway, the reason they're not ranked yet, is because I was doing a study on horses with a particular "thing."

So, aside from the next step being to rank them, what else do you guys suggest? Where to go from here? Or is it already a "trash it" factor? Or do I bet all of 'em for the ROI?

Cheers heaps,

AP

[tdthomas]

I agree with Dick Schmidt on this. I thoroughly handicap every horse in the race before deciding which ones are non-contenders.

[JustMissed]

:) Did anyone ever hear of a quick scan method to review the pp's. I read on the net a little about it. Your suppose to be able to take a red pencil and go though the whole race in about two miniutes.

I have been using ProCaps and handicapping the top five or six power rated horses with any top five or six on the speed report that are not included in on the ability report. I like the ProCaps because of the pace charts but would like to train myself just to use the pp's for now as I am a new player.

If anyone knows about this red scan method, would you please post how it works.

Thanks

[Dave Schwartz]

AussiePlayer,

You have stated ROIs of +12 and +16% with a significant number of plays. In my opinion, you need nothing else. Just bet them all.

In the US we would NEVER see such a significantly sized group of horses in each race that would produce such numbers.

But a caveat: If your sample is small, test against a larger one before assuming that you got it nailed. (Or, of course, just give it a try with small dollars and let time tell.)

Dave

[anotherdave]

Originally posted by JustMissed
:) Did anyone ever hear of a quick scan method to review the pp's. I read on the net a little about it. Your suppose to be able to take a red pencil and go though the whole race in about two miniutes.

I have been using ProCaps and handicapping the top five or six power rated horses with any top five or six on the speed report that are not included in on the ability report. I like the ProCaps because of the pace charts but would like to train myself just to use the pp's for now as I am a new player.

If anyone knows about this red scan method, would you please post how it works.

Thanks

Sounds like Jim Lehane's technique. He calls it the red scan qualifying technique.

http://www.free-horseracing-info.com/handicapping-cal.html

AD

[andicap]

I used to like his stuff and then he went out on his own started charging for his picks and lost a lot (at least according to reports we were getting back on this board).
Has anyone been following his exploits more recently.


ag

[rrbauer]

I like to leave them all in, first cut. It just gives me a keener sense of the race.

Although, more recently, I've been tossing Zippy Chippy in my early routine!

:)

[Myhorse1_X]

Anotherdave:

Have you had experience in using the Red Scan Qualifying method, and if you have, could you tell me about how many contenders it narrows the field down to, and how many winners are in those contenders.

Thanks

MyHorse1

[anotherdave]

Originally posted by Myhorse1
Anotherdave:

Have you had experience in using the Red Scan Qualifying method, and if you have, could you tell me about how many contenders it narrows the field down to, and how many winners are in those contenders.

Thanks

MyHorse1

Sorry, not really. I just remembered seeing it mentioned on Jim's site.

AD

[Tom]

Personally, I like to narrow down the field to 4-5-6 horses, depending on the field size. I think leaving them all in muddies up the water and makes it harder for a clear model to be seen.
I rate them all (so to speak, actually, HTRT rates them all) and then I make a first cut. I use non-velocity factors at this stage. After I have the field down, I use the velocity numbers to make my selections.

[Tom]

Originally posted by Aussieplayer
Myhorse1: That's the same method that Gordon Pine mentions. (He averages, but adding has same affect).

Tom: Where you say, "Ignore the last two races, take the best of the remaining Beyers, use the top three. You now have 3-6 contenders. " Do you mean the best remaining from anywhere in the pp's? Or do you mean using the third last race?
>>>>>The first three come from the top two pacelines only. Then look at the rest of the pacelines for each horse and get the top three from pacelines 3-10. The same horse might be picked in both groups. So the forst pass, you consider two lines, the next time, you pass by those and use the other 8,

Tom (again): In the last method you mention, how many lengths (roughly) is 10 Beyers? From memory, about 5?
I think using "within10 points in last 2 races" would make most of the field a contender?
>>>> About 4 lengths in sprint, 5 in routes. No, a lot of times I end up with only a few contenders left. If you have the DRF for Finger Lakes for Belmont day, 6/8, check out this method....I forgot how good it was.

Cheers to you!
Cheers
AP :D

[Aussieplayer]

Dave: Thanks. It is a small sample, yes. I know the "how many is enough" is a never ending question/debate, and I've read that maths stuff that gave me a headache (lol), but I would really like to ask what some of the real-world players would feel comfortable with? Assuming the results stay (for the refined method) roughly similar (40% hit rate from 3.5 per race (which is approx. 28% of field), all races with hopefully the same ROI, or even half that ROI of 16%). Dick, when would you start betting? 200 races? 500 races? 1 000 races?

As an aside, I realise more and more (for me), that starting with a small roll is the way to go. I just can't escape those thoughts I mentioned to (Dick) a couple of threads back - you never know when something is random luck!! So, if you take a small roll & profit take along the way, you (I) can never get hurt. "Feel" hurt is probably the more correct/honest term!! lol
That's my take on money management anyway!!

Back to the contender topic! I should clarify as some of you might have thoughts on this:
The "method" I'm talking about doesn't actually have a fixed amount of horses per race. In theory it can be 0, right through to the whole field. In practise, very few are zero (so I shouldn't have said all races , but it is at least 95% of races!).
A few have 1 contender. (This is the refined method I'm referring to). Most races sit comfortably at 2-5 horses. There are a few however, that have half the field!! I've been testing the whole group, so haven't paid attention to price etc. But obviously on the occasion of having 8 horses to bet, the fav. is no doubt one of them & there are going to be at least a couple of horses that you are NOT going to make any money on!!

Any thoughts on "what to do?" The 2 options that spring to my mind are:
Option 1: Drop the lowest prices that are going to stop you making a profit. e.g. If 5 horses, drop the favourite, and see if you can back the rest. If not, drop the next shortest and see if you can back the longest three. Kinda reminicient of the Key to the Mint.
Option 2: (prefer this one) - hedge on the shorties to break even and flat bet the rest. Obviously if you have 3 horses & all are 3/1 or more you are going to bet all 3. If one is the even money fav and the other two are decent odds, you would put hlaf your stake on the fav. to break even and split the other hlaf between the other two good odds horses.

Another option that I REALLY like is what Dave mentioned recently: back the best of the two favs. & the best of the rest. I would have to obviously research this, and I will have to rank them and research, but I very STRONGLY suspect that some of the good priced horses (20/1 etc) will be missed.
That was the interesting thing with betting the "group." No ranking meant the 20/1 shots could be bet.

I never considered this to be viable, and wouldn't have thought of a non-ranking system for contenders normally. Was just researching a couple of factors and found the group to be profitable as a whole.

Thanks to all,
AP

[Aussieplayer]

Tom: Thanks for the clarification. All three methods are interesting. (I find contender selection methods interesting lately, lol). I think your third method: best last race Beyer from similar distance, same surface within 30 days & anything within 10 points is the most "valid" (for want of a better word), as it opens it up to have anywhere from 0 contenders to the whole field. This is a much better approach (IMO) than one that "forces" a fixed amount of contenders.
I think you could even refine your initial requirement to demand that it also be a "good" or "competitive" race - however you want to define it!!

Cheers
AP

[Derek2U]

hehe ... i will re type what i typed maybe 1 month ago:
you guys are very very close 2 being declared InsaNE.
There's no animosity in my saying this, so kinda chill, but many
of you think so classic yesterday, and yet, kinda funnier yet,
forget the earlier advice of guys like Ainslie, preferring instead
Quinn/Brohammer/Mitchell/Cramer/nameless others with databases bigger than China's population. Or with Brohammers'
equations Bubbling-Over w/ adjustments ... hehe .. what silly stuff
why do us guys fall in love w/Horses/gambling/Ego/science .....
Money2. It's a total guy thing and i'm happy that way 2. Guys +
A stud like WE .... not a bad bar partner.

[Aussieplayer]

"anyrate Consider the Horse... "

Derek2U, that's what I am trying to do. How do you apply this?

Cheers
AP

[Derek2U]

yeh thats the Bigquestion thats 4 sure but i cant stop thinking
about horses differnt than numbers. they got 4 feet to hurt we
guys got only 2 and u know what happens when a Bball player
hurts his tiny toe... 1st , the guy athlete is able to SaY in clear
voice ,,, "heY TonY i Hurt My bitty digit" rub it / call my lawyer also.
hehe .. the horses mostly hurt without sayin & so form is #1
factor with the Exact distance today. i can get rid of more than 1/2
of the field in less than 10 mins ... almost 95% accurate at this
stage. hehe i only play NY tracks like i said b4. i think form is
the single unifying factor... although u can b more forgiving if
you see a layoff dixie brass on a wet track .. hehe .. much like
overlooking lots of circumstances .. which is why all them kiddy
studies on form look so banal .. like never wager on a pisces
horse with a capricorn rising.. heheh

[Tom]

If I get a lot of contenders then I assume that speed is not really the dominant factor in the race and look at other things. If I only get a couple or three, then I assume speed will be important. One think I always do for exactas, though, is look at the noncontenders....may recent Beyers are bad, but back a ways they are as good or better than todays 10 point range. ThenI look for things to indicate maybe a run back to the old form, things like 3rd after a layoff, a class drop, claim to a better barn, equpment change, etc. At Finger Lakes, you have to pay attention to small details. Almost all the horses have run far better in the past than they have lately and big improvment is always possible.

[Dick Schmidt]

Aussie Guy,

I'll try to answer some of your questions.

Assuming the results stay (for the refined method) roughly similar (40% hit rate from 3.5 per race (which is approx. 28% of field), all races with hopefully the same ROI, or even half that ROI of 16%). Dick, when would you start betting? 200 races? 500 races? 1 000 races?

I believe in starting early and betting often. Once I find something positive, I jump right in. I'd be betting this after 100 races.


As an aside, I realise more and more (for me), that starting with a small roll is the way to go. I just can't escape those thoughts I mentioned to (Dick) a couple of threads back - you never know when something is random luck!! So, if you take a small roll & profit take along the way, you (I) can never get hurt. "Feel" hurt is probably the more correct/honest term!! lol
That's my take on money management anyway!!

I agree that starting with a small bank is best. Let the pari-mutuel machines grade your system for you. I always start with $200, as that is what I started with originally back in about 1982. I have never added to that original bankroll.

Another option that I REALLY like is what Dave mentioned recently: back the best of the two favs. & the best of the rest. I would have to obviously research this, and I will have to rank them and research, but I very STRONGLY suspect that some of the good priced horses (20/1 etc) will be missed.
That was the interesting thing with betting the "group." No ranking meant the 20/1 shots could be bet.

I've never much favored letting the public make my betting decisions for me. I do check for minimum odds, but after that I bet the horses that I think are the two (or three, or sometimes one) best horses in the race. I have found that I do best when I ignore reality and treat handicapping as an exact science.

Dick

[alyingthief]

a method using beyers to select contenders was offered by Scott (he of how will your horse run today fame) in one of the early handicapping symposia offered in vegas. you take the top 5 last out beyers, and the top 5 best beyers showing: you end up usually with 5 to 7 contenders in full fields (due to overlap). the winner appears in these numbers over 90% of the time. at least, as Scott told it.

as to the comment that your method should do the picking, betting certain horses at certain tracks on the basis of numbers, be it pace, class, or speed, will eat your bankroll and you for lunch. you really have to know your tracks to pick up these nuances, and it's hard to believe a method can ignore such knowledge...and pay for itself.