Jerry,

Thanks for the response and your patience but I am going to test that patience a bit more.


Just to preface my next remarks, let me assure you that I use your download product and the #'s work fine. I have made $ with your #'s which is the ultimate bottom line to this game IMO. In my line of work, I am known as a fairly (actually my constituents would say overbearingly) analytical guy so your methodology intrigues me.
Point 1) I understand
Point 2) I understand
Point 3) If nothing was done to the Bel, then how does Sarava compare to ...say...AP Indy, Point Given, Easy Goer, Secretariat and Commendable?
Point 4) I understand your take but I am not trying to make any race fit anything or "look better". I just noticed that the early sprints had a lot of horses hitting new tops or jumping way up. You indicated that you used "Basically" the same variant for those races. Would I be correct in assuming, then, that you make slightly different variants for each race??
If that is the case, do you use a preset par time for each type of race (Gordon Pine or your own) and, then calculate individual variants and then try to tie them altogether. Or do you (as David P represented) project a par for each race based on past performance and, then, make a variant based on those projections?

That question is the essence of my confusion. I had always used (of course, this was some 25 years ago) pars based on past races in each class of race.....compared all the races after the fact and gave the track a plus or minus variant for the day and adjusted #'s accordingly. (I did separate turf, dirt and routes and any meaningful weather related change would naturally screw up the works).
I did not massage the #'s in Beyeresque fashion. The number was the number was the number. Crude maybe...imperfect no doubt but they provided a discernable edge at the venues I played at that time.

From what I can ascertain, you really don't use pars. I say this because a few months back, (in my only other posting) I asked you what par you utilized for MCL's at SA. I asked this for the purpose of gauging first time starter's chances in these races. You indicated that the par actually changes for each race? I admit that, to me, that makes little sense.

For example, if a MCL 32,000 6f has a par of 15 which is equal to 1:12 for the distance, then it is useful to know if those who have already had a few efforts have ever broken par or exhibit a pattern that threatens to do such. No?

If a bunch of plodders are entered, then I can upgrade the chances of the First timers and so on.--- (I am saving your winning numbers in these categories and am slowly evolving my own par for this purpose)

If one projects each race and uses that as "quasi par", then if would follow that the variants would be somewhat 'customized'.(as they are based on an interpretation rather than a database of past races of the same class at the same track---or the traditional "par" as I knew it)) It would also follow that each # was somewhat 'customized'. This, of course, would also imply that future 'quasi pars' would be based upon these customized numbers and so on and so on.

Jerry, as I prefaced, whichever way you do your numbers, they work. I realize you are probably shaking your head and laughing at where this lunatic is going with all this but I am trying to grasp the basics of how you make figures as I feel that will better enable me to use them effectively.

Thanks again

Bob