Okay, now that the noise has died down, let's get to the important stuff. Friedman has left an awful lot on the table for me, and I won't get to all of it, but I'm going to get to a lot of it.

Before I start, a couple of things. When I began this site I established it as an open forum, for several reasons-- to get what the Ragozin operatives say at racetracks (where there is no one to contradict them) out in the open, to show that (unlike Friedman) I have nothing to hide, to explain how figures are made, and because in general I'm a big believer in "sunlight being the best disinfectant". At this point the first goal has been pretty much established, at least until large numbers of Ragozin players who don't know any better move from hard copy to the internet, and the negatives are now outweighing the positives-- I've had to spend a lot of time I can't spare recently explaining the difference between a dictionary and a thesaurus, and the static in general has distracted the focus away from where it needs to be. This has created a significant tactical advantage for Friedman, and now that he has abandoned any pretense of taking "the high road", my previous approach is a luxury I can't afford.

So we are changing the policy. Unlike Friedman, I will continue to leave up and address questions and negative comments that actually are pertinent to evaluating the data. But general detritus will be removed, at my discretion, and even with the questions I do address, I won't engage in endless back and forth. So you have seen the last of the no-content twins (Jim and Soup), at least for a while. The latter (Dave), who thinks clever is better than smart, had flipped out again-- he was sending posts from around the world to avoid being blocked, just so that he could fill our board with trash and push the significant stuff down where no one would read it. He appears to have stopped, which is just as well, since we were about to have a word with his isp in Chicago.

Briefly, two other things. First, someone recently said (again) that we should get off the question of accuracy of the figures, and back to handicapping questions. I will point out again that this is the most fundamental of all handicapping questions-- if the data you are using is not accurate, the rest of the decisions you make can't be. Also, for obvious (commercial) reasons, it is the question that concerns me most, and it's my site.

Secondly, for those of you that may not have been following what has been going on, you should read Friedman's post ("Re: Churchill Downs 4/28/99 Race 1", posted 1/31/03 their site), and my reply ("Friedman fires smoking gun, shoots foot", 2/03/03, here). For those that have already read those, I will point out again that FRIEDMAN HAS YET TO EXPLAIN HOW THEY CREATED A VARIANT (AND THEREFORE A FIGURE) FOR THAT RACE.

1- In Friedman's post, he mentioned figures for 4 1/2 furlong races at Keenland. Since he had confused CD and Kee at other points in the post I assumed this was more of the same, then realized that Ragozin is in fact making "figures" for these races (Awsome Humor being one example). When we started making figures we did as well, until I realized that:

a) The horses running in those races never set foot on the backstretch that older horses race over in sprints. They instead race about an eigth of a mile over a chute, that older horses (anything older than an April 2yo) never sets foot on.
b) The chute is downhill, the backstretch is not (which is crazy, by the way).
c) The chute enters ON the turn. The 2yo's don't circle the entire turn.

Now, there are ways to make figures for 4 1/2 furlong races. At many tracks older horses run that distance, enabling us to get a firm fix at those places, since those horses have extensive figure histories to use as a basis for comparison (the basis for all speed figures, in one way or another). Additionally, most tracks have one of a few standard configurations that enable us to extrapolate using the relationships once they are established between sprint distances IF the circumstances are truly comparable-- meaning, among other things, that the run-ups are not significantly different. Note that this is a completely different question than that of tying different races together within one day for variant purposes, which brings a lot of other questions into play, as I have pointed out in other posts.

The situation with the 4 1/2 furlong races at Keenland, as you can see, has absolutely nothing to do with any of the situations described above. Older horses never run over that course, and there is no other track where older horses run over an identical (or even comparable) course, even leaving aside the questions of whether the soil composition is the same in that chute and the main track, whether it gets the same track maintenance, and how many other tracks of same configuration would be enough to draw conclusions from.

Which brings us to Friedman's point that the figures hold up in light of subsequent events. This is ridiculous-- these are April 2yo's, and developing very quickly. Whoops! That's not right, is it-- SOME OF THEM are developing very quickly, others are already precocious, some are in between. And of course, since these races take place at the beginning of the horses' careers, there is the question of fitness-- they are all making their first or second lifetime starts, and apt to SOMETIMES move forward explosively for that reason.

All of which is the reason that when it comes to establishing variant par levels, Ragozin and TG use only older horses-- we don't even use 3yo's, let alone April 2yo's. Even to use 3yo's would mean setting up a sliding set of levels for each month of the year, and the results would be wildly erratic within that, but there is NO WAY to draw conclusions about April 2yo numbers based on later figures of the same horses.

You want to know how they came up with those Keenland figures? They made them up, based on ridiculous assumptions-- everything else is BS. They MIGHT be of some use in comparing one Kee horse to another, ASSUMING run-up relationships and track maintenance never change. But they are of NO use comparing a horse who ran at Kee to one who ran somewhere else. Bottom line, they didn't want to leave a box, so they did not. So much for "objective" figures.

2- Once again, this is Friedman's post from 6/25/02: "It is true that we only slide our variant when the physical resiliency of the track changes-- a practice that produces accurate, objective numbers".

This is a quote from Friedman last week, with emphasis added by me:

"...neither that race, or for that matter the whole (day)was an easy example of variant making. There were several tough decisions that we STRUGGLED with using the best AVAILABLE data, experience and skill we could COME UP WITH and gave out the numbers that all factors considered LOOKED best.... NO GUTS, NO GLORY".

This is, at last, an accurate description of how figures are made, although it says nothing about the crazy assumptions and other problems inherent in Ragozin's particular methodology. But that aside, it also has NOTHING to do with measuring physical resilience or creating objective figures-- there is no way to measure that and no such figures, and if there were you sure as hell wouldn't need guts to make them.

I'm going to add this as well, on the question of experience-- at this point I've almost certainly done more projection style track days than anyone who ever lived, about as many as Ragozin and Friedman combined, and about 4 times as many as Friedman alone. The kid's still wet behind the ears.

3- Turning to the 00 Wood, which Friedman has now reposted-- there is so much here that I almost don't know where to start, and I won't get to all of it. Those who want to get the full history of this should read "Figure Making Methodology", 5/2/00 this site, and reposted recently, as well as Friedman's recent post about the race. The gist of the matter is that it rained during the card that day, the track was sealed after a couple of races, and unsealed right before the Wood, and I cut the the Wood loose from the other races. Ragozin gave the race faster numbers than I did-- several horses jumped to new tops-- and when I said they got it wrong, Friedman explained how they made the figures by saying they tied the race to the surrounding races and that it was correct to do so, despite the work done to the track (similar to that done on the Chilukki day), which is in fact the only reason you would give out those numbers.

Cut to this past spring. In the dust-up that followed the Preakness, Friedman finally conceded that track maintenance (never mind the rain) COULD change track speed, which was obvious to everyone all along. If it CAN affect track speed, you MUST split the races-- there is no longer any validity in tying them together. If they come up with the same or similar variant that's fine, but THERE IS NO REASON TO MAKE THE ASSUMPTION THEY WILL ONCE YOU CONCEDE THAT THE WORK CAN CHANGE THINGS. That was the significance of Friedman's post last spring, not whether track work affected the last race at Pimlico that day, although that is how the question came up. I have no reason to believe work was done that day.

So, as of last spring, the original rationale for those Wood numbers was out the window.

Now, a couple of days ago, Friedman changed his tune, saying that the only race they tied the Wood to was the only other race after they re-opened the track that day, which by the way was not a 2 turn race. This is not a good idea for other apple/orange reasons, but makes more sense than what Friedman originally said they did, which several others who posted on both boards at that time said was a good idea.

Okay Len, here's the problem-- if you are now saying that it was not correct to tie the 00 Wood to the races run over the sealed track, how is it okay to tie Chilukki's debut figure, run over a sealed track in the opening race on the card, to the races that followed it, run after the track was unsealed? It's enough to make a guy think the decisions are arbitrary, the opposite of objective.

Len's other point is that subsequent events vindicated those Wood numbers. I have already gone into the problem with that-- 3yo's are developing, so they will run faster as the year goes on. Additionally, if there is one group of 3yo's you will use to make future figures with it will be top class distance horses, so the results will be to some degree self fulfilling, no matter who is making the numbers. IN FACT, THE CLASSIC EXAMPLE CAME OUT OF THIS VERY RACE. Those of you who were using Ragozin figures at the time might recall that there was exactly one horse in the 00 Derby who paired his top in that race (these are spring 3yo's mind you). That horse was Fusiachi Pegasis, and the number he paired up was the disputed Wood figure. Does anyone really think that if FP had gone into the race 2 points slower Ragozin would have done the Derby the same way? Hey, it's only the Derby. Not that important.

But the part of Friedman's recent post that leaves me a little breathless is his attempt at revisionist history. Subsequent events did NOT make Ragozin's figure look good. Friedman did not like Fusiachi Pegasus or Aptitude (the first two finishers in the Wood) in the Derby BECAUSE THEY HAD JUMPED SO MUCH IN THE WOOD-- in fact, in his seminar Friedman gave FP just a 10% chance to repeat his top. Since he was off a smaller forward move on TG, I said in our seminar that while FP was an underlay at a short price, he was 40% to repeat or go forward, and our play in he race, on the record, was Aptitude, who finished second, and would have won if Solis had stayed inside. We still have the "multi-media" seminar, so if someone wants to hear it we will put it up again.

This post is long enough, so I will hold off on the trainer stat stuff for now. But I do want to say something to "Charlie Varick", who made the cute comments on the Ragozin board saying the reason I post about Ragozin's numbers is that I "define myself" in terms of him:

Charlie, on-line sales for the first 6 weeks this year are up 45% from the same period last year (despite the site being down on Sunshine Millions Day), and last year as a whole the site was up 40% from the year before. In the first 6 weeks this year 436 different people have purchased data on-line, up from 334 last year. Additionally, hardcopy sales at sites around the country were up 15% last year, despite us only giving 3 seminars in the last 3 years-- one at Gulfstream a year ago that started my friendship with Florida Jim, and 2 last spring in California. Also,DRF stopped hosting our website last March, so we no longer get all that print advertising.

So, why are sales exploding? Itis true that our strategic alliances (TVG, DRF, NYCOTB etc.) probably play a part. But the primary marketing outlet for us is this board. You may not be reading this stuff carefully, but a lot of other people are, and a lot of them are now ex-Ragozin customers



TGJB