Rules Rules Rules

[RichieP]

There are NOT any rules that can't be broken or amended for a specfic race or situation that arises. Therefore there are NO rules.

THIS race at hand:
u "feel" line 4 is needed USE line 4 heck consider line 5 if your gut is telling you that

u "feel" a tier 4 or 5 guy ( in a tight mix) has some positive things goin for him and maybe a couple ranked better have a negative or 2 shoooot bet the lower tier guy

u "feel" u want to use this line from 200 300 400 days ago USE IT

It has taken MONTHS of being ALONE working races live to realize this in spades.

There is nothing etched in stone when it comes to working a race choosing lines hiding wager decisions etc etc


THERE ARE NO RULES

Merry Christmas
Richie
PS - Brooks and Mrs. B Thanx so much for the card! u kinda shook me up with that! I love you!

[partsnut]

Ritchie,

You make a good point.

Prime examples: Varigate: It's wrong as many times as it's right.
" " : Pace Lines: Very subjective, what happened in the past may
never happen again. We can only guess :confused:

" " : Models: They actually prove nothing and lose money.
Just another ploy for the software sellers but it keeps you busy.

Answer: Go with what works for you and stick with it. If it doesn't work find a new pastime or in some cases an occupation.

[shoeless]

Richie and Bill I understand what you are saying about rules but you should have some type of guidelines for selecting pacelines.As for models Bill I used to take your approach and guys on here would tell me you have to keep them and then I never did.But Pino made up one for me on excell at first I resisted but thanks to his constant badgering I started to and Im glad I did.For instance Im using Synergism 2 and they have something called a DR rating.Out of 20 races going 6.5 and 7.0 in claming races for winners 17 out of the 20 have been no worse than 3rd in this ranking.If I didnt keep a model I wouldnt have known this.Anyway just my opinion.Richie glad to see you posting again even if it's only once in awhile.Jeff

[partsnut]

Jeff, I am not try to be condescending but if your winner comes from the the top 3, you have approximately a 67% chance of picking the wrong horse to win. If you had 2 choices and played them both you might have a shot at a positive ROI.

[Binder]

Great point
everytime the subject comes up I suggest the same thing..
20 race cycles bet two horses to win

Jeff the DR ranking? Your tracking how often the# 1 Dr.
horse is the winner? How about how often the # 2 Dr horse wins ?
Then what is your win percent if you bet Both your # 1 and #2
DR ?
Lets say you do 40 races from CT at 7.0
Your # 1 Dr wins 20 and your # 2 Dr wins 10
Now you have 30 out of 40 Dr ranked 1-2
I bet if you notice all the DR winners
the # 2 ranked DR is also good but if you only bet 1 horse you
may be blinded by the loss
just a suggestion

Bill

[shoeless]

You all make very good points.I do keep track of the 2nd as well and there are a lot of times ties.I basically try to use the DR as a contender guideline like you would do with total energy in Val.Basically they are probably just about the same but more refined with Val .

[socantra]

Jeff, I am not try to be condescending but if your winner comes from the the top 3, you have approximately a 67% chance of picking the wrong horse to win. If you had 2 choices and played them both you might have a shot at a positive ROI.

Your 67% figure might have some validity if you are talking about blindly making your choices from one readout, but if your model holds at 85% winners among your top 3 contenders, you have gained a significant advantage from which to start your selection process.

Dick...

[tompkins]

like all of horse racing: there are no rules only different interpretations

[partsnut]

Your 67% figure might have some validity if you are talking about blindly making your choices from one readout, but if your model holds at 85% winners among your top 3 contenders, you have gained a significant advantage from which to start your selection process..

To my mind, and my thinking could be off base, each race sets up differently.
Naturally, our picks are base on what we deem to be the contenders.
It doesn't matter which program you use, although I will state that some are better then others.
The question that I have is why would a model of past races have any bearing on what will occur in the present. To me it's almost like saying that if I kept a model of the craps table which contained 10000 rolls that it would help me predict the outcome. You can't predict the outcome because rolls of the dice are random occurences. I will say that a horse race has more variables then craps.
There are too many variables in horse racing that can effect the outcome of a single event.

I am a firm believer in probabilities, especially when then are in my favor.
Most winners come occur from Validator when the Sartin odds are 7-2 and below. Normally, I look at 3-1 and below that will go off at a price.
The way the race sets up is also significant (Early or Late). I look at Energy and Entropy as well.

This type of model using 5 contenders in Validator might be worth your time.

The bottom line is ROI - It's not about how many winners you pick but how much money you make.

[socantra]

If you keep a model of a crap table, it will show you that some numbers come up more often than others and what those numbers are. This will do you absolutely no good at craps because the occurrence of the various numbers is already built into the odds and the house still has the advantage.

In horse racing, the top three public choices win in the neighborhood of 67% (I can't remember the exact figure). If you can find a model that indicates a factor produces the winner in the top three 85% of the time, you've just gained a nearly 20% edge on the public choices and it doesn't depend on the odds.

Yes, models are based on the past, but so are past performances. Most methods of handicapping are dependent on models of the past. With the exception of physicality handicapping, we are all pretty dependent on the past being a reasonable indicator of the future. If not, we might as well throw darts at the racing form.

Dick...

[partsnut]

socantra:In horse racing, the top three public choices win in the neighborhood of 67% (I can't remember the exact figure). If you can find a model that indicates a factor produces the winner in the top three 85% of the time, you've just gained a nearly 20% edge on the public choices and it doesn't depend on the odds.

If you have found a factor that produces 85% winners from your top 3 contenders, your time was well spent. The rest should be easy. You may have found the key to the bank. I was wondering if you can elborate on this factor, and how you derived it and how you are using it.. Is this for all classes, surfaces and distances?
What program or programs are you using?

[socantra]

Nope, I haven't found the key to the bank and if I had, I would be unlikely to share it with you or anyone else. Bank keys are very rare indeed and sharing them usually makes the bank change the lock.

The 85% figure I was referring to was shoeless post about a model showing 17 of 20 winners at a particular distance and (I assume) a particular track. If you want to know more about the particular factor, you need to talk to him. I believe it was something in one of Bob Purdy's old Sartin Synergism programs.

I was largely reacting to your dismissal of modeling as a waste of time. There are many little niche plays like that out there and you can find them quite often if you do dilligent modeling. It is a lot of work, but nothing comes free. I have yet to see one that applies at all distances and surfaces. I suspect that modeling is a waste of time if you are looking for such a 'key to the bank'.

I use Validator primarily and I am participating in the testing of a non-commercial Sartin style program with extensive modeling features. It makes the modeling easier, but its still a lot of work. I'll sing its praises if the author ever decides to go public.

Dick...

[gl45]

I must agree with Dick statement. Modeling do require a lot of work. Each track might produce a different winner and place model. What works at Tampa might be out of wack at Philly. I play one track and only two distance 6.5 and 7 furlong, same class and conditions, I don't play NW's. Using the same software's factors, did produce a complete different winner and place models at both distances.

6.5 furlong top 3 rankings:
===================
Early/Presser win 56%
SP factor win 73%
Fx factor win 73%
Wx factor win 82%

7 furlong top 3 rankings:
==================
Sustain win 45%
Fw factor win 80%
Sp factor win 80%
TT factor win 73%

as we can see, same track 2 different distances, 2 different models.
This is not a key to the bank, but it make easier to for me to determine the most likely contenders.
GL45

[RichieP]

FWIW I have bet this track every day since it opened

6f dirt - ROI minus (-)23 % 25 races
8.5 dirt- ROI 232 % - 17 races
8.5 turf- ROI 145 % - 7 races
7 dirt - ROI minus (-) 24% 13 races


think I am looking forward to the next 6 and 7 f races?? :eek:

Happy New Year
Richie

[Rxlawyer]

To All:

I think both arguments have some validity.

Models are useful, but only to a point. Sometimes they do point to some factor that is particularly strong at a given track and at a given distance. If you find such a situation then, by all means, take advantage.

From my experience models are actually used in the reverse manner as described here. Specifically, to my way of thinking, models should be used to exclude horses in a race and not include them. For example, if your model shows that a race can not be won by a horse with a MED% over 70.00 (this info comes from your model) then you exclude all horses in the given race with MED% over 70. Will a horse with a MED% over 70.00 ever win? Yes. When the track has a significant bias toward early, by fluke, or when every horse in the race has a MED% over 70.00.

Also, the problem with models is that they show what already happened, but only in a "gross" kind of way. What I mean by "gross" is that the models only take so many factors in to consideration. A model can not tell you that the speed horse broke from post 9 and was blocked by a lesser speed horse from post 8 causing the factor of early speed to fail this specific time. Models can't tell you that the jockey took the horse 8 wide on the turn.

Yes, models are good for however many factors are measured. But no one I know of models the 1000 factors that are actually present in a race. And no one models the interaction of those factors in a race. Do you have separate models for races where there were 3 speed horses in the race as compared to races where there were 5 speed horses in the race? These are the issues that limit the value of models. Yet, models are still useful for elimination purposes, at least in my view.

Gary

[tompkins]

every race is different. rules are then only RELATIVE

[admin]

Thanks Gary

"From my experience models are actually used in the reverse manner as described here. Specifically, to my way of thinking, models should be used to exclude horses in a race and not include them. For example, if your model shows that a race can not be won by a horse with a MED% over 70.00 (this info comes from your model) then you exclude all horses in the given race with MED% over 70. Will a horse with a MED% over 70.00 ever win? Yes. When the track has a significant bias toward early, by fluke, or when every horse in the race has a MED% over 70.00."

I was working a race with Doc once. The winner paid like $17.00
He had all the readouts I sent him I was lookling at my computer
we were talking on the phone
(thats how it worked back then)
He asked me why I didn't have the winner
I said "Because it is over 70 Med energy"
Doc got real quite then sain something like
Thats Not what I said or mean.. What about The total energy ???
I do better like you suggest using models and record keeping as a eliminator rather then a posative betting method

Bill

[tompkins]

a couple of weeks ago an Aqueduct maiden projected SEVERAL top lines into a route with ALL the lines at, or over 70% median. After a double take, I realized this one was the ONLY speed of the race. He won wire to wire having NO pace pressure: EACH case requires a decision based upon the HOLITICS of today's race and NO predetermined database values OTHER than as a guideline.

[tompkins]

these types abound in maiden races as this shows 2nd race Aqueudt 3/29
Pgm Horse Jockey Win Place Show
7 Margarita Rita Pablo Morales 4.70 2.90 2.40
3 Trick a Cat Fernando Jara 4.30 2.60
8 Two Skats Pablo Fragoso 2.60

[admin]

Thanks Tompkins for that thoughful post
I really can see clearly how you could not miss that winner

Here is a race I bet today 3/29 Oaklawn
5.5 furlongs race 10

Here are my Pace lines

[admin]

after hides of multible lines and horses with over top 5 Line score
and Total Energy -and ties- I could easily eliminate the # 8 Texas4 but it is top 5 Total Energy so I will leave it in
This is the BL/BL

The very first thing notice is the 6/1 ML for the 7 horse
and the the ML fav is ranked lower in tier 4

[admin]

Biggest stick early ! Oh NO !!! at 5.5 furlongs
trouble ? wire to wire? short price ?

But here is the key for me
Horse# 5 is 6th Total energy
The program already told me the 7 was better than the 5
on BL/BL

My next post will show the Med energy readouts
Check out Horse 5

[admin]

Here is the power of the program
looking at this screen and the L/E seems like th 5 will be loose on the lead
from multable readouts

Results are ok
Small problem the 7 was bet down to
$7.40 Using Richie P's "Soccer Guy"
no emotion betting method I will take it


Results
7 Running Brave $7.40 $ 3.60 $2.80
3 Mossant $3.40 $3.20
1 Etbaur $6.20

[tompkins]

why would that one be loose on the lead as it's early potential is only SECOND and the total energy to way down in 6th?

[admin]

Exactly my point

Posting screen shots and dicussing them sure clears the fog