Apv And Class Importance

[froggy]

Storm,

On screen #8 where the print out shows L I R instead of LSD, it is only a programming error. The line separating LSD and R didn't get properly spaced during programming and the SD is eliminated when you print it out.
It is just moved over that's all.
LSD is an acronym for "Long Shot Detector" and is the total of the numbers on screen #5.

As for your APV/CLASS figures they are only a guide. Most bombs seem to come from low numbers as in your race.

When you printed out screen #8 you saw an earnings per start chart.

To save time I sometimes just use the figures in the program and eliminate the lowest 40%. Ten horses in the race I throw out the lowest four.
I try to balance the current year with sprint races or route races depending what kind of race I am doing. Your own judgement must be used and as you learn the better you will get. Look at Bill's post today on the race at Tampa. I belive you will find it very informative.

I find Turf races to be very sensitive to APV

I think Bill has a video that explains how he does his APV. I'm sure he'll point you in the right. direction.
It is also explained in the old and new Yellow Manuals.

You don't learn unless you ask questions. We were all new at one time.
I know I am still learning.

Good Luck

Froggy

[froggy]

Jim,

"Per I" refers to screen #6 which is the Perceptor screen.
This is where I each horse is ranked on each of his individual read outs
and a number/percentage is assigned to each read out and then totaled up for a final ranking
A very good screen, very useful.

Froggy

[Ted Craven]

Jim, Froggy,

Just to elaborate a bit on Perceptor 1, this is a way of portraying the 7 Primary Line Score factors in a way which shows the degree of difference between different horses for the same factor, rather than only showing rank differences. In Speculator, the highest value for, say EPR, is shown as 100% and ranked #1. The 2nd ranked value may be 95%. Meanwhile for, say LPR, the 2nd ranked value may be 99.5%, yet both 2nd ranks are equally ranked #2. However, the 99.5% rank#2 is stronger than the 95.5% rank #2.

The Perceptor 1 Total, is simply a sum of the percentages, which is then more sensitive to the relative strength of the component values than a simple sum of ranks. The P1 number itself has no intrinsic value, but the Difference between the numbers and their rank is the value. In most cases, the resulting rank of Perceptor 1 Total agrees pretty much with the traditional Primary Line Score sum of ranks, but where it differs, you might consider investigating why and perhaps see that one horse has a significant gap in an impactful factor (i.e. for today's race) over its nearest competitor, which simple rank values might hide.

A note: in RDSS, Perceptor 1 is labeled as Primary Line Score screen, and the highest value for a given factor is shown as 0.0% rather than 100%. Since the Perceptor 1 Total number has no intrinsic value, I dispensed with it, showing simply the Difference between best P1 Total, plus the rank of those Differences. Same thing, hopefully a bit more clear.

Ted

[Ted Craven]

Froggy,

Re APV and Class, if you search for posts by Admin within the past 7-10 days, you will see one where he posts a nice video explaining in detail how he computes APV. Also recently, Admin posted a link to an excerpt from the Yellow Manual discussing APV. Further, if you consult Folow Up Issue #5, Page 7, (e.g. in the Pace and Cap Library), Dick Schmidt discusses a revised calculation for Class/APV. I'm not sure which methods were implemented in which various classical software (Engen, Energy, Phase III).

Ted

[froggy]

Ted,

Thanks for the complete explanation.

Froggy

[socantra]

Originally Posted by stormbringer

.......another Question: I Have Been Looking Back At Some Of The Other People's Apv/class Numbers Some In The Don't Look Like The One From The Phase3 Program I'm Using. Could You Or Somebody The And I Hate To Say This Word "rules" To Doing Apv/class Numbers. It Was My Understanding That A Horse Need At Least 7 Starts?

APV/CLASS numbers were simply an attempt to mine the earnings box for useful information. There were many different formulas used in the programs. You would probaly be hard pressed to find any two programs that used exactly the same formulas. All of them combined an average purse calculation, average earnings per start and some sort of win/in the money calculation in an attempt to tell you how well the horse has been running and in what sort of company.

The Schmidt article Ted mentioned is a very good background. The calculation can be skewed by slots income as it was by state bred programs, but the idea is that better horses follow the money and run more consistantly.

Seven races is good. I believe Schmidt mentioned six in the current year. Either way, you want as much data you can get on how the horse has been running recently. Its not a stand alone and is imperfect like any other single tool. Its another corrallary and you're the one who has to decide what its worth to you based on your own records. Some people believe in it very strongly and some don't use it at all.

Dick

[Houndog]

Socantra, appreciate your explanation on the different ways APV and Class can be calculated. I agree with you that some people use them and some don't. In the past I have been in the group that does not use them.

With Admin's posts on the subject I really think I need to take a second look at APV and Class. Thanks all for the comments.

[froggy]

Dog,

After you have lost a race go back to the section of SPEC160 about running lines an uncheck the pacelines of the horses that were are in the bottom half of the earnings per start.
Then go back to BL/BL. I did that a dozen times or so and it humbled me.

Froggy

[socantra]

Originally Posted by froggy

Dog,

After you have lost a race go back to the section of SPEC160 about running lines an uncheck the pacelines of the horses that were are in the bottom half of the earnings per start.
Then go back to BL/BL. I did that a dozen times or so and it humbled me.

Froggy

To be fair, also go back on the races you won. I suspect that you will find like I did when I tried a hard and fast rule using earnings per start that I did indeed improve my winning percentage, but reduced my ROI considerably. Horses with poor rankings on earnings per start win races every day, just like horses with poor jockeys and poor trainers.

I think some sort of APV/Class figure can be a very valuable corrallary, much more so than earnings per start by itself, but trying to apply a simple rule to its use will throw out many good priced winners, like the one at Lone Star that started this thread.

Dick

[froggy]

Dick,

I agree.

Froggy