Quote:
Originally Posted by Mark
Peterman,
It makes a great deal of sense. But from being right on to way off, you never know. So I don't use them. I do use ITV however.
In my view final time is completely reliant on early pace. I don't know the percentage but I believe it is a good 50%. You don't get fast final times without fast early fractions. If you have a bunch of SPs and S in a race or slow Pressers, nobody wants the lead. You see it in turf routes all the time. You see Pressers who can finish against a 1:11 but some slow E or EP grabs the lead and they play follow the leader through 1:12. There is a limit to how fast a horse can run the final fraction, they are flesh and blood.
So it makes absolute no sense to me to make final time variants that don't in some fashion take into account the early pace and it makes less sense to apply a final time variant to early fractions. It's nonsense. I don't know why some really smart guy hasn't figured a way to do this. Hell, it has been done this way since early TPR days. Pizzolla showed how he adjusted DRF speed ratings by 2nd calls off his fulcrum pace. If I was 20 years younger I might make a stab at it, but I have gotten to lazy and so I just don't use them.
|
Mark,
I think you are correct about never knowing just how accurate DTVs can be.
And I absolutely agree that trying to determine track variants based on just final time alone is not good enough, and absurd without context of early fractions.
However, its clearly better than no assessment of track surface speed; and the reasons why no one has figured out how to do it
(what I highlighted in bold) is two-fold at minimum imo:
(1) it's almost impossible to do without handicapping each race (before and after the running) which involves more than just crunching numbers on a spreadsheet.
Even with enough man-power or software creating knowledge, extremely time consuming. and,
(2) because it seems every one who has attempted this process only views it through either: final time 1st, or just velocity alone.
Class, current form and position at each call are still very hard to piece together consistently for virtually all of us.
I've only been back into the game for just a few months, but I've been attempting to piece this all together for only a few weeks.
A sometimes frustrating process that pans out for me at the windows from time to time, which I ultimately am convinced is solvable as far as
discovering a more efficient way of determining track surface variants.
Then I ran across this thread. Hopefully, I can be one of those who become smart enough to bring the pieces together.