Trackmaster revises Speed and Class

[Ray2000]

Anyone buying TrackMaster Programs are probably aware of this but in case you missed it, TM has changed the way it calculates Speed and Class Ratings.


http://blog.trackmaster.com/



Rather a cryptic description of what they're doing, but I don't think I like the changes to Class Ratings. Age? History?

This alternate method of class ratings is no longer based solely on the speed ratings of the horses in the race (as is our standard method), but uses other factors like age, earnings and other horse related statistics to determine the class.

[Pacingguy]

Originally Posted by Ray2000

Anyone buying TrackMaster Programs are probably aware of this but in case you missed it, TM has changed the way it calculates Speed and Class Ratings.


http://blog.trackmaster.com/



Rather a cryptic description of what they're doing, but I don't think I like the changes to Class Ratings. Age? History?

This alternate way to come up with class ratings is going to occur in very few races. For example, races at Goshen Historic Track where there is no-wagering. You get a lot of horses that race in NYFS racces so you may want to dismiss these races, but everyone once in a while, you will find a decent horse who has been racing in the pari-mutuels. In the past, the class rating would have been NA, but with the alternate class rating you will have a value. A horse in that race goes on to the Meadowlands or other pari-mutuel track; don't you want to get a feel for the caliber of horse?

This will occur in about 1% of the races. It gives you something to go by. What they should do is denote it is an alternate class rating; perhaps a dififerent color or box the number.

[pandy]

I don't think Trackmaster has actually changed the way they do the regular Class Ratings. They created a different way of making a class rating for horses that have no class rating. The email they sent out was confusing, but from what I read, when a horse doesn't have a class rating they came up with a way to give it a class rating, rather than say no rating available. I do not believe the regular ratings changed at all.

[Ray2000]

Pandy, Pacingguy

After the 5th re-read of their notice ..I believe you're right. I asked the question on their BLog anyway.

[markgoldie]

I have been using TM numbers for years in my handicapping. The changes they are making are essentially irrelevant to users of the numbers. First, changing their trotting scale to be more in line with their pacing scale does little to help the handicapper. Why? Because even though their trotting numbers were more than a "few" points higher to similar pacing numbers, I see no reason why anyone would be comparing pacers to trotters when handicapping an individual race.

In isolated cases there might be a small advantage to trainers and/or owners who were looking to claim a horse. For example, if you had $10,000 to spend on a claim, accurate cross numbers might tell you which $10,000 animal was a better buy relative to the class in which it was racing. But even that is a stretch, since the animal's relative advantage to the field in which it races is more significant anyway.

As far as adding extrapolations for fair races, I think we can already doubt the accuracy of these numbers, even before they are published. This will all be the worst kind of guesswork on their part because without an extended meet, the assumptions involving a daily track variant will be basically impossible. So assumptions based on age, sex, and past history will simply lead to more confusion, not less- the point being that literal use of these numbers will only get you into trouble.

As I have stated in the past, the thing they need to address more than anything they are doing here is the track-to-track inaccuracies of the numbers. This could use immediate correction if they are serious about making their product better to the casual user. The best example of these anomalies (although there are many), is the Maywood-Balmoral disrepancies. In this case, the Maywood numbers are consistently about five full points higher than corrsponding Balmoral numbers. This inaccuracy is all the more unforgiveable because so many of the horses race back and forth between the two tracks. So its not as if TM can use lack of cross-reference as an excuse. In fact, there is no excuse other than the inability of TM to analyze their own product.

While I'm on this rant, they should also immediately drop any and all adjustments to the numbers based on post position. The reason for this is rather simple: The effects of a particular post position are only relevant in the context of how the race was actually run. And, of course, such analysis is far beyond the ability of a computer program to assess on a one-size-fits-all basis. For example, a post position #8 is normally a hinderance to performance. But if the horse is a sole leaver and falls into the two hole in a slow-paced event, was he more hindered by post than the #6 who was wrangled off the gate and winds up racing seventh and must attempt to make headway into a fast last half? The point is that post positions are relevant only as a function of the particular race pattern and therefore cannot be assessed in such a simplistic way.

It's nice to know that the powers at TM are attempting to improve their product. But maybe they should consult some players who actually bet real money before deciding what changes are most urgent.

[LottaKash]

Originally Posted by markgoldie

IIt's nice to know that the powers at TM are attempting to improve their product. But maybe they should consult some players who actually bet real money before deciding what changes are most urgent.

Nice slant on this issue Mark....I agree as well, especially the track to track adjustments, they stink on ice...And the off track ratings leave much to be desired as well, they are inflated if you ask me, and not a good number to put your money on, that is....

STill, in more recent days, I find myself using the TM class and speed ratings, less and less, as time goes by....I have found other ways to express the same thing, I guess....

best,