Philadelphia Park Data for Decision Making

[Bill V.]

Data includes all non maiden races for the fall meet starting 8/30
to 9/7

For each distance, I have made separate data sheets for 3YO (if any so far)
3 + and 3+ Fillies and Mares

Not too many races yet as they are running lots of 5 furlong turf races
and I don't record anything under 5.5.
Also they are running many 7.5 Turf races with no fractions

The races are listed in order of the race date and race #
for a track profile summary

for your perusal my fellow Sartin Meteorologist
Good Skill
Its a great day at PHA

Bill V

[mikesal57]

nice going Bill...this is a lot of work but its necessary to WIN!!

one more thing if you can ..add a column and just put the class of that race..no biggie , but it will help
Mike

[SilentRun]

Bill,

Thanks for the effort you are taking to model, I play Philly fairly regularly and the data
you present is a complement to what I model.

Ernie

[Bill V.]

Thanks Mike and Ernie

i am not to happy with the format. I'd rather do it the way Doc says
separating by Total Energy High Low and Average and distance

From Follow Up #32

[mikesal57]

Bill you were headed in the right direction with your profiles..

Yes its a lot of work in the beginning but it will pay off!!!

lets give an example...

Track profile at one mile 10claimers......5-3-2-1 (position only)

horse a- 2-2-1-1 total energy 161 2-1 odds
horse b- 3-2-1-1 total energy 163 3-1 "
horse c- 6-3-1-1 total energy 159 6-1 "

Who would you choose?

[SilentRun]

I have found that in sprints up to 6.5F TE is not as significant the fourth factor
(% of energy E/L). This is not only true for PHA but across a significant number
of tracks. To lock onto the early contenders I have also found that
E/L, F1, SC , E/EP and EPR to be more reliable than TE for sprints only.
For each playable sprint I now set up a decision table (3 deep)
using the above indicators and I have been having good results.
Many of these races are so similiar that I note them as PR (Pattern Recognition).

Ernie

[SilentRun]

This race is an example from the above post. Horses 10 and 7 were the top
2 contenders for E/L, F1, SC and EPR. They were alone on the front end and finished 10/7 and I was on it.
Notice the top TE horse finished 3rd and the 2nd best TE horse ran out.

Ernie

[For The Lead]

Originally Posted by Bill V.

Thanks Mike and Ernie

i am not to happy with the format. I'd rather do it the way Doc says
separating by Total Energy High Low and Average and distance

From Follow Up #32
Attachment 20560

I can understand the reason you weren't happy with what you were doing. Simply put, it was wrong.

Now that you posted what "Doc" said to do, maybe you could explain it.
In #1, he says to separate total energy by class, with an explanation of possible high total energy classes, but in #3 he says you SHOULD NOT keep this information separated by class.
In #2, he talks about low total energy and that low total energy is usually represented by 3 year old fillies and males. Naturally there is no such thing as a 3 year old "mare", but not matter, again he goes on to say in #3 that "sex" SHOULD NOT be used.

Do you see how this can be confusing?

If you DIDN'T separate by class, how would you know the high total energy races would be Allowance, Stakes and/or Handicaps?

If you DIDN'T separate by sex, how would you know the low total energy races would be from 3 year old females?

[For The Lead]

Originally Posted by SilentRun

I have found that in sprints up to 6.5F TE is not as significant the fourth factor
(% of energy E/L). This is not only true for PHA but across a significant number
of tracks. To lock onto the early contenders I have also found that
E/L, F1, SC , E/EP and EPR to be more reliable than TE for sprints only.
For each playable sprint I now set up a decision table (3 deep)
using the above indicators and I have been having good results.
Many of these races are so similiar that I note them as PR (Pattern Recognition).

Ernie

Nice job. Just one note. I think you confusing "decision models" with "track profiles". The information you get (and use) from the bl/bl is what gets used in a "decision model" not a track profile.
Other than that, I agree with what you wrote and it is apparent that you are doing some homework.

[SilentRun]

Originally Posted by For The Lead

Nice job. Just one note. I think you confusing "decision models" with "track profiles". The information you get (and use) from the bl/bl is what gets used in a "decision model" not a track profile.
Other than that, I agree with what you wrote and it is apparent that you are doing some homework.

Your right about that, thanks for your comments.
Then I guess you can say I am using the track profile to make a decision.

I also hit the 2nd and 5th races using the track profile. In the 2nd race the 7 & 9 horses were the front runners both were the top 2 for F1, SC, EPR
and at nice odds. So I went with the predominately early contenders and the 7 won paying $18 and change. Notice the 7 is the lowest in TE.

In the 6th race the 8 won paying 10.60 the track profile was:

F1: 11, 8

SC: 8,7

E/EP: 7,8

EPR: 8, 7

TT: 7,4,8

So there is my track profile (decision) The 8 and 7 dominates.
And the exacta was 8/7 paid $33.00.
I am having very good results amoungst many tracks using this method.

Ernie