Re: Par Times: New York Pars--TG
Posted by Jerry B on September 27, 1999 at 16:28:06:
In Reply to: Re: Par Times: New York Pars posted by HP on September 27, 1999 at 13:57:06:
: : To statement about there being a linear relationship and the earlier comment re the Turfway stake; my point is that if you take a large enough sample say 1000 or 2000 races, and look at a horses TG number (I eliminate the variable of ground in this because I ONLY look) at 1W trips) and then do the adjustment for weight (i.e. 1lb = 1/5 pt) and regress it against the horses Beyer figure you will get a formula to translate one to the other. That doesn't mean that you could take a Beyer figure multiply by say x and add say y and get the exact TG figure (even ignoring groundloss) because different figure makers will come up with different estimates of the day's track variant. But, this is whether the law of large numbers comes in...the differences in varaints bewteen the two figure makers will average out to 0 over a large enough sample (assuming they are both compentent) leaving you with that one equation:
: : AGAIN this only applies on average and not in EACH AND EVERY INSTANCE which is why you note discrepencies relative to the Turfway race. Assuming you accounted for weight that difference will be in the daily track variant assigned. By way of validation, I challenge someone with enough back sheets to calculate NYRA pars by class for the past year. I'd bet there much closer to the Pars I have 'derived' than the universal pars presented.
: : And to open up another can of worms..."Just how do you use these Pars?" There is a fairly wide body of evidence that demonstrates that ANY figures are less predictive when the whole field is literally SUB-PAR. In other words if the par for a group of horses is 10 and no entrant has broken 12 it is less likely that final time will be the major handicapping factor at play here. For whatever reason in fields of sub-par horses factors like pace and bias play a much more prominent role. If you don't believe my read some of Quirin or Quinn's stuff.
: : Anyway, I think I've gotten in enough trouble already. It's too bad because I was just trying to help people. I've got no axe to grind in this one way or the other.
: If you look at BJ's example (the Turfway race), this kind of discrepancy between Beyers and TG happens ALL THE TIME, so I don't think this 'law of large numbers' applies and balances it out. Your equation is crude at best given all the factors you are leaving out (ie., wind) and adjusting (1w trips). As for the other can of worms, Quinn, etc. notwithstanding, how about ONE example of YOU using these par times. For Christ's sake, ONE! You spend all this time on this and you can't produce ONE example! Anyone! Now you are just trying to 'help people'. Before, if we didn't accept your theory of the 'linear relationship between Beyer and TG numbers we just didn't understand the math'. Well, demonstrate YOUR understanding! If you want to help, HELP US LEARN, show just ONE example. I guess this is just your little secret. You are not 'getting in trouble'. I for one am serious about this game and I want to learn as much as I can, but that doesn't mean that what you say is going to be accepted as fact. By the way, for what it's worth I think bj's postings have been right on. Please Ian don't play the persecuted martyr.
TG--I put up my few thoughts on the subject of pars up on 9/26 (Odds Must Be Crazy and Par Times), so I obviously think that you guys are making a lot out of not so much. But whatever, everybody calm down, okay? You all seem like intelligent people, so leave the feuding to me and Ragozin.
Len, you ignorant slut... TGJB