Come on, Len. Knock it off.

[TGJB]

Just got back from playing St. Andrews (43 on the front, as good as I can play, blew up on the back, although the heat and 30 mph wind didn\'t help any).

More annoying stuff to deal with, mostly from Friedman.

1-- The study Len quotes is very old, from way back when we had hand written sheets. It was done by a guy with a Vegas newsletter. It featured several mechanically done studies (best last number, etc.), and Ragozin did better than we did, although as I remember, worse than others did (specifically, the old DRF figures based on track records, I believe). The guy who did the study, Jim Bayle, said himself that he was not satisfied with it and did not think it was meaningful. So he later gathered data from Ragozin, us, and several other figure makers, for six months, at something like 10 circuits. After doing all that he did not run the studies, because it was interfering with his betting (!!!).

2-- Let me get this straight, Len. You don\'t want to do Jimbo\'s study because it is purely mechanical and doesn\'t take into account patterns, but you quote the results of a decade-plus old study that was done purely mechanically and doesn\'t take into account patterns. Do I have that right?

I\'ll let the rest of Len\'s nonsense go. But I\'ll say that while we both use patterns, those will wash out over a long run-- meaning, in one race it will hurt one of us that we are not using patterns, in another it will hurt the other-- and, the patterns are always a matter of opinion and percentages (see the Thoro-Patterns). Numbers are numbers, and in the long run, faster horses will beat slower horses. There are lots of ways to do a study, but Jimbo\'s works for me, although I would do the scoring a little differently. I\'m willing to do it, there is really no reason that Len would not if he has confidence in his figures, except for marketing reasons.

But also-- Indulto, you are such a putz. In case you haven\'t noticed, the sheer number of posts about me and TG on the Rag board is staggering-- like, maybe half the posts there, more than that recently. Maybe 5% of the ones on this board are about Friedman or Ragozin, or even JJ. In that recent exchange, there were at least 5 times as many posts attacking me and calling me names as those going in the opposite direction. Did you even notice?

Jimbo can take care of himself (and you, without breaking a sweat), but one more thing you might want to think through: he is proposing an objective, purely mechanical test. While it may or may not be indicative, there is no way, especially since he wants to recruit a Ragozin guy to do it with him, that it can be biased in my favor. So how in that weak brain of yours do you come up with the idea that it is some nefarious plot? Objective tests of data are bad, and insidious TG doings? Do you read the stuff you write?

[kev]

what\'s the big deal with this : \"2-- Let me get this straight, Len. You don\'t want to do Jimbo\'s study because it is purely mechanical and doesn\'t take into account patterns, but you quote the results of a decade-plus old study that was done purely mechanically and doesn\'t take into account patterns. Do I have that right?\"

He said no to doing it, but if he wanted to really know about doing something like it, there\'s already been a study done. Let me ask you this........Why would even any of you guys ( Len--Jerry ) want to do a test like this anyhow. This is not how you teach your people to use the product, what if one or the other did real bad, what would it do?? Nothing, that\'s why one of you would say what I just said, that\'s not how we use our product. The big studys should be done in pattern\'s and bounces and good stuff like that. Jerry you down for something like that anyways--the best fig. out of the last three crap?? Hell what if they threw in beyers number\'s and they did better, ouch.

[jimbo66]

Kev,

I think I was pretty clear on the other board in that I mentioned more than once that I was very open to suggestions on how to do the study.  There has to be a way to compare the products, I would think that any user of either product would realize that.  Of course, I know that patterns are part of how most of us use the products, but how exactly would that be factored into the study, without adding any level of subjectivity, which ruins the effectiveness of it.  

It is also pretty clear with Friedman\'s dumb comments \"Sheets have no competition\", \"We don\'t do marketing\", etc.etc, that he has no interest in a comparison of the products.  Last time I checked, we lived in the U.S. and it is a capitalistic society, all products have competition, even the \"Sheets\".  The fact that he claims he does no marketing, yet apparently supplies his product free of charge to guys like Frankel in exchange for public testimonials, shows he is being more than a little disingenuous.  By the way, I think what he does with Frankel and a few others is good solid marketing, but to make stupid statements that are misleading on his board is not honest.  

[richiebee]

TGJB-- Congrats on the 43. Sounds like you \"bounced\" on the back 9, but of course the wind should be taken into account.

Jimbo-- Since these various tools are used in different ways by different people, my belief is that it would be very difficult to rate these products(Beyer, Sheets, TGs) in any scientific way.

Possibly the most efficient way to undertake the comparison you mentioned would be for a single unbiased handicapper to use both products during a meet, alternate days or maybe alternate weeks. Considering the fact that Rags and TG are making figures for tracks all over the country all year round, a sample based on results from one short meet wouldn\'t be too convincing.

To this neophyte, the tools produced by The Guru Jerry Brown and the self proclaimed \"Father\" of all speed figures seem so similar in terms of methodology-identifying patterns, seeking animals ready to peak, avoiding animals ready to bounce.

These are both good products, IMO. I feel comfortable preferring to use one without disparaging the other. The long running feud between Rags and TGJB, the colorful vituperations of some of the posters on the other board, these are nothing more than sometimes entertaining distractions.  

[kev]

That\'s cool. I don\'t know if there could be a way of doing a pattern study vs. both companies. Take ten Rag. user\'s and ten TG user\'s and there would be so many ways people would look at each horses lines. I have tired this along time ago with both products, mostly on BC day. I could do it, because I know what pattern\'s look good to me, but they might have looked different to someone else. Why did all of this come about??

[jimbo66]

Kev,

I gave my honest motive on the other board.  I am a gambler and am open to any theories or products that increase my ROI.  I think that comparing the products would prove interesting.  I did some less formal comparisons this summer with a few Rags users and it peaked my interest.  The two companies disagree alot.  It would be worth it to me to spend some time and money to compare them.

Patterns are of course important, but you must realize how different the patterns for the same horse can look on each of the products?  Since you have used both products before at the same time, you would know that on one product,  a horse could be sitting on 0-2 and be ready to \"x\", while on the other he may have paired up his top and be ready to move forward.  That is why to me, the simple mathematical formula I proposed would provide a way to see whose figures are more accurate, despite the lack of pattern reading in the study.  Reading and interpreting patterns is more style than substance, knowing which horses are the fastest before the race is more important to me.  If I know that, I can start from there and figure it they can run back to it.  But like I said in my post, that was just my opinion, I am open to alternative ways to measure the data.

RichieBee,

Respectfully disagree with your post.  Sheets and Thorographs are not used in different ways.  They are used in very similar ways.  For this reason, the quality of the data is key.  How we as handicappers interpret that, is our own issue.   But the figures have to be as accurate as possible.  

[Chuckles_the_Clown2]

I dont how you intend to do an objective study.

If one set of figures more accurately depicts the truth regarding the efforts run, the pattern reads with that set should result in more reliable projections regarding the coming effort. However, the subjective human interpretations exist as well. Will the horse bounce, regress marginally, pair or move forward?

I don\'t think the Rags are nearly as good as TFigs, but even if they agreed to a study, I just don\'t think its possible to use a study to determine how accurate the projections were.

For horse racing there is only one machine that is nimble enough to take all the data and vary the weight assigned to it depending upon the scenario. Its the human mind.

I didn\'t see the proposed study. If someone has it in text please print it. I\'d like to review it.

CtC

[sighthound]

I\'m no statistics major, and I didn\'t see Jim\'s proposal for the study.  I\'m a lurker, and a very minor player, but this certainly piques my interest. My suggestion:

10 players, competently trained to each system, and with their own individual backgrounds of experience handicapping - obviously not novices.  5 use TG, 5 sheets. Heck, you could even add 5 more using Beyers.  

Each handicaps all races on the card for one or two designated days a week, at a designated meet, throughout a 6-month period.  At 13 weeks (halfway) into the study, they simply switch products and continue.  

They handicap their usual way, watching simulcast, tote, track conditions, whatever additional adjuncts they normally use with the exception of the other product.

They have to pick three horses:  designated most likely winner, and two additional placers in no particular order.  This has to be submitted somewhere (time/date stamped e-mail) before each race.  

Players have to be dedicated enough to not be sloppy on the assigned handicapping day - offer the player at the end with the largest straight win percent $500 or something.

Will yield objective analysis by many different measures: % wins, % place/show, breakdown by type of race (turf, sprint, route, maiden, age, etc), \"quality\" of track/meet/horses/trainers, payoffs for variety of bet types placed on those three horses can be created in the lab later with ROI for each type of bet based upon real payoffs calculated, etc.

The working hypothesis is that there is no difference in results between products.

Alot of the \"statistical nightmare\" worries of individual user differences and style of handicapping or betting inherently and simply taken care of by the double-blind study design.

I\'d be willing to bet that the win percentage for both products will be very similar if the sample size is large enough. In fact, I\'d bet they will be so close you won\'t be able to tell whether the difference can be attributed to the quality of the product or randomness. The ROI result will at least partly be a function of whose figures are contributing more money to the pools - also not a matter of quality.

I\'ve done some smaller studies on different sets of speed figures that I thought were of high quality (never TG and RAG though) and they always produced very similar net results despite the fact that they often disagreed on who the fastest horse was. It was enough to convince me that very small differences in the figures don\'t matter much to the win percentage and that it\'s highly likely that no one has a monopoly on mistake free figures when we are talking about a length here or there.

I think the real trick is those occasional races where there is a huge discrepancy due to either methodology or interpretation. IMO, those don\'t show up often enough to infuence the aggregate results heavily, but if you have a preference for some reason, IMO that should probably decide whose figures you use.

[miff]

The problem with any type of comparison is the person interpreting the figs. I\'m friends with many RAG users and we compare figs and opinions all the time. Sometimes they are right and sometimes I\'m right. I win more often than my RAG friends BUT I\'m  more into \"spot\"shots and they are into spreading and stabbing in the exotics and multi wagers. Not a fair comparison.

One way to compare would be to get a few experienced sheet players from both camps and give them both sets of data. Let them handicap a card with TG and make their selections. Next,let them handicap the same card using Rags and make their selections.I think after 30 cards in should be fairly evident how successful each person is with both sets of data.

Of course, TG and RAGS could put together a serious, legal, \"buy in\" type contest where everyone could put there serious money where their big...err opinion is.The results of such a contest may give some indication as to which data is more solid.  On the other hand, it may only prove who is the best interpreter.

[horsegoer]

Listen JB. Listen I may have to settle the score once and for all aand buy your selections everyday for a couple of weeks and pick my selections using the oh so very ACCURATE SHEETS. Yes I know you don\'t make the selections but it is your product and you obviously have faith in him/her. I am sure I could just track the results of your selection and that would be just as good. Better yet how about me and you have a contest. I\'ll put up a thousand bucks, chump change to you thanks to your endless promoting.

[TGJB]

Beau-- I have a better idea. Talk Friedman into taking me up on one of my ten or so challenges to a long-term handicapping contest, and you and I can bet on that. And more than a thousand.

[Chuckles_the_Clown2]

Assuming Friedman and the Rags powers that be are not going to acquiesce to a \"study\", why don\'t you guys accept Horsegoers challenge for a period of two weeks?

Equal Number of Raggies vs. equal number of TFig devotees.

They can pick their Rags winners and TGraph users can pick their TFig winners.

Saratoga and Belmont, winners for the card selected by pre offtime postings. One winner selected per race scores two points for a win, one point for a place. (Reliability is the issue, so the 100 dollar winner shouldn\'t skew the contest/study.)

Its more contest than study, but contest is the only way to prove the superiority of the figure in the long run.

One player might get lucky. But with 10 on each side that probability lessens. Total wins, total points determine winning team and figures.

Kinda a group contest

That might be an interesting handicapping contest but it wouldn\'t say much about the figures. It would say something about the skill of the handicappers. The only way to measure the figures is to remove all human/subjective elements from the selections - like best figure last time out, best of last 2, average of last 2 etc....

By the way, that wouldn\'t be true if you were comparing figures other than RAG and TG because Beyer, Bris etc... don\'t incorporate ground loss and weight.

If you compared anyone other than TG and Rags, you would be measuring both the quality and whether or not adding those things into the figure is a net plus or minus whithout really being able to tell whether quality or content was the reason for the performance.  

[TGJB]

Kev-- let me make the point clearly. The single most important handicapping decision you make is about what data you use to make your decisions, since all decisions that follow are based on that one. That doesn\'t just go for which sheets to use, it applies to whether one should use sheets at all, and for that matter whether one should look for patterns.

With that in mind, trying to determine which data is better should be of paramount importance to all sheet players, as it is to Jimbo-- especially if you are looking for patterns, where those small difference can make a BIG difference.

I like the last 3 numbers, toss the worst, average the other 2 approach, in terms of giving an overall power rating. It\'s not perfect, Len and I would undoubtedly bet against the top rated horse often. But as I said, over a large sampling those pattern questions would wash out, from a statistical point of view, leaving it to a simple question of who is fastest, next fastest, etc.. It eliminates all judgment questions and arguments.

I would score it this way-- rate all the horses in every race that way, for both of us, in order. If the winning horse was number 2 on theirs and 4 on ours, they win. You could also do other studies-- top rated horse for both in every race, in terms of win % and ROI, for example. And if Len wants to go farther back with the figures, you could take best 3 of last 5 or something like that, or best top, although I think the original idea is most reflective of horses\' current abilty.