BC -- Rags vs. Thoro

[Shrek]

Interesting post, JB, about Arezi.  My friend uses Rags, and I noticed the same thing.  But in comparing sheets, one could just as easily pick out e.g.\'s where your numbers seem absurd.  Like the idea that Dynever went forward in the Pa Derby off the W Va Derby.  I mean, he loses a neck in 1:46 and change to a horse who comes back and runs within a length of one of the top 3 year olds in the country; then gets beat a country mile in Pa race.  Beyers say he went back 30 pts, Rags say he bounced 6 pts; you say he went FORWARD!

Everyone blows numbers.  The q over long haul is whose are more accurate.  I tend to think yours are and that\'s why I use them.

But if you\'re going to raise the issue, let\'s do it fairly.  Compare each of the BC races and see who comes up stronger.  For e.g., since you brought up Distaff, I\'d say Take Charge Lady would be an interesting test case -- looks like a very live play on TG, a complete throwout on Rags who hasn\'t gotten back to her 3 year old top.  In contrast, Elloluv looks pretty good on Rags, not much on TG.  We could do this race by race if you want.  Then we\'d have something to really talk about

[kev]

Yes, I bought both sets, the main one I will be using this year is Rag\'s. I used TG last year, I got TG set this year for the Euro Horses. So if i have time i should pick who I like on both, sometimes if a horse looks good on both it\'s time to load up on it.

[TGJB]

Responding to a few points at once--

1) I certainly want to compare the results of the BC in terms of us and Rag, as always, but I don\'t want to be tied to Shrek\'s  interpretation of my figures, especially since my own are a matter of public record before the fact. In the end, all we can do is make the differences public, and leave everyone to draw their own conclusions.

2) There are a few exceptions (War Emblem\'s pair up of the big jump in last year\'s Illinois Derby comes to mind, and Storm Flag Flying returning to the disputed big # in the BC), but in general the result of one race is seldom even extremely strong evidence, let alone proof, that one figure on one horse is correct-- that is why I added the disclaimer about my comments being seen as an endorsement or knock for the BC.

I can think of a couple of examples from the last two BC\'s that I pointed out in advance where there were big differences that made certain horses look a lot better on ours than Ragozin-- Tiznow and Volponi. Hopefully I\'ll have time by Tuesday or Wednesday to pick out some differences on this year\'s card that lead to horses looking a lot different-- right now I\'m swamped. I was up until 2:00 A.M. last night writing the BC comments, and only got through 3 races.

3) If you haven\'t already, check out the PA. and W. Va. Derby #\'s posted below on this site, and look how they fit together, and read my response to Michael Downey about doing the Pa. day. Figures can fit so well together that you know they are right no matter what happens in the future (I am talking about a small sampling, one race, not about figures overall), and certainly short term results are unlikely to disprove a figure (although this year\'s Derby did a hell of a job backing up what I said about Ragozin\'s Wood #\'s being off by a point). They will more often \"prove\" a figure, if horses come out of the race and pair up what seemed like an unlikely figure, like War Emblem in the Il. Derby last year.

If you look at the Pa. Derby figures you will see they are rock solid. And as it happens, Grand Hombre came back to pair up that figure exactly in the Indiana Derby, meaning he has now run that exact figure 3 times in a row. His performance in Pa. looked better than it was (inside trip, getting weight), his Hoosier effort worse (wide trip, high weight), and I for one would have very much liked to see him in the Classic at 40-1 or so.

Those who don\'t have access to accurate information about ground loss and weight corrections have problems dealing with things like Dynever\'s effort at Philly-- when you made all the corrections, he only raced 5 lengths worse than the winner, not the 12 he was beaten. Which means if he had been at level weights and had the same trip as Grand Hombre he would finished second in the race, and no one would be disputing the figure. The problem he had was that at the weights he actually had no shot in Pa., which everyone using TG knew going in. And certainly Dynever\'s forward move last time doesn\'t make the figure and great TG pattern look any worse.

4) The Calder sprint day was a very tough one, and I added a little to the 3yo race too, although not as much as Beyer. In general the Beyer guys have a more common sense (meaning correct) approach to doing variants than Ragozin, although they are significantly disadvantaged in other areas (weight, wind, ground, graph, and last time I looked beaten length corrections, although that was a long time ago-- TimeForm has the same problem, and it causes distortion and figure making problems when a winner wins by a lot, like Shake You Down did that day).

But Andy\'s answer about Valid Video was a little simplistic. The key thing when you look at a race with a winner whom you are considering giving a big number to-- Shake You Down, Valid Video, Grand Hombre, War Emblem in that overnight handicap last year where he broke through-- is to look at context, not just that horse. That means the rest of the horses in that race, and to a degree the rest of the similar races on the card (which goes to the question of what \"similar\" means, which I\'ve gone into in great length on this site before).

Anyway, I\'ll have a couple more pop quizzes later today if I have time.

[TGJB]

Kev-- you used our BC sheets LAST YEAR and aren\'t using them this year? Holy smokes.

[kev]

Well, it\'s like this. I seem to do about the same on both RAG and TG turf races. The higher stakes horses is where I like to use the Rags and the lower clamier levels I find your product better. I can\'t get use to your neg. numbers, maybe i\'ll wait till you rasie the scale up some. Like i said i bought your BC product this year also. But i will being my betting picks threw Rag\'s this year.

>Those who don\'t have access to accurate information about ground loss and weight corrections have problems dealing with things like Dynever\'s effort at Philly-- when you made all the corrections, he only raced 5 lengths worse than the winner, not the 12 he was beaten.<

The problem is greater than that. It wasn\'t presented clearly in the past.

Dynever\'s Beyer figure for the W.V. Derby  \"adjusted for weight and groud lost\" is still much bigger than than his PA Derby Beyer figure adjusted for weight and ground loss. (114 vs. 88 = 26 pts naked = about 16 lengths on the Beyer Scale).  A pickup of 5 for the PA Derby pounds and being wider does not account for a 26 point switch or 16 lengths on the Beyer scale.  

The W. Derby race was also definitely a difficult figure to make because it started raining just prior to that race. So immediately it was suspect for everyone - including Beyer\'s 114!!!!

However, Dyenver then came back with an effort in the Med Cup in Beyer terms adjusted for weight and ground loss that was more similar to his WV Derby figure. (107 naked, again wide and carrying 3 extra pounds). (I would subjectively also argue that he closed into the hot part of race after 2 very slow quarters - which makes the effort even better)

So some people have simply questioned whether perhaps that 114 Beyer - which is a very big figure for a 3 year old at that time of year and much better than the 2s or so you were giving him) was actually correct.

That would mean he was better than he looked going into the PA Derby on your figures, didn\'t like the mud, and then ran back closer to his better figure in the Med Cup as opposed to improving.

Further evidence about the WV Derby came from the very good race that Soto ran against TMW.

Similarly, many of the horses that Dynever was supposedly similar to going into the PA Derby did not have terrific credentials at that time, have since finished up the track in their next race, are nowhere to be found now, and Dynever is heading to the BC Classic.

Grand Hombre earning extremely impressive TGs relative to just \"good\" Beyers was less of an issue.  

IMO, Beyer\'s, and apparently Ragozin\'s, the WV Derby figure is still open to question.

In case anyone is curious.

The Indiana Derby received a Beyer figure of 103.

Grand Hombre received a Beyer Figure of 101 carrying 124 pounds. He earned a 108 in the PA Derby carrying 114 pounds.

In addition, he lost a lot of ground in the Indiana Derby worth approximately 6-8 Beyer points if one were inclined to make such adjustments.

[TGJB]

You are killing me with this stuff. I don\'t have the time.

I ILLUSTRATE some of the problems with Ragozin figures for those who haven\'t made their own figures (which as far as I can tell includes all Raggies who have ever posted on either site with possibly one exception, and also apparently includes you), or haven\'t used sophisticated ones, by using common sense analogies, like comparing a performance in a nw1 with a G1.

I MAKE figures using very in depth and sophisticated data including weight corrections, ground loss, beaten length corrections, the relationships between horses that are created by that data within a race, a day, etc., and the past histories of those same horses using the same kinds of highly evolved data, relationships etc. I don\'t look at whether a horse like Soto ran a \"good\" race against another \"top\" horse on another day when MAKING the figures. WHAT WE DO HERE IS INFINITELY MORE SOPHISTICATED THAN THAT.

I EXPLAIN how some figures were arrived at to those who have made their own figures and/or are sophisticated in their understanding of figures (of which there are LOTS who post or read this board), by posting some races so they can see how the relationships played out with all the data input, and look to see what the alternative approaches wre in doing the race/day.

Anyone who is sophisticated in the making or use of this data (meaning not you, although some of your last post is an improvement, and I\'m about to get to that stuff) knows that because Beyer does not use weight or ground his data can only be used as a rough guide, especially in cases where there are significant weight or ground loss differences. This is not a knock on Beyer-- as I have said before, they are a tremendously useful tool for the small, recreational player, especially at the price. But I use them as examples only in the most general, rough guide sense-- if there is a dispute and the figure based on all the stuff I mentioned above is solid, we are right, and they are wrong, if only because their figure makers are at such a disadvantage in terms of accurate data to use to make decisions. (And that would be true vs. Ragozin as well if the latter didn\'t follow such ridiculous, dogmatic rules, and used common sense instead).

Dynever raced 5 paths wider at Philly than he did at Mnr, and carried 5 more pounds. Those differences alone account for about 12 Beyer points. Additionally, as you mentioned, it rained before the Mnr race, and it appears Beyer (and possibly Ragozin) failed to know it and cut the race loose, or disregarded it. If Beyer had cut it loose and done the race based on the previous figures of the horses, he would have given it a MUCH slower figure. They blew it.

The point you fail to get, and everyone else who has made figures does get, is that the way those two races came together, with horses coming out of different races on different circuits especially, MEANS they are right.

Anyone who hasn\'t already should go into the archives and look at those two races. Beyond that I will be deleting any more on this-- it\'s BC week, and I ain\'t got the time.

[beyerguy]

One other note about the W Va Derby...the pace of that race was very, very slow for the track conditions.  If the final Beyer was a 114, the pace number was about a 91.  So it would certainly explain the horses flying home to such a big number.