Steven Roman - Dosage Index, Leaving the game

[Fager Fan]

[QUOTE=PaceAdvantage;2142097]Posts have been deleted and edited, and a penalty has been assessed to Fager Fan.

I told you guys I wasn't screwing around anymore. You want to act like kids, you get treated as such.

Make it civil in the horse racing section please.[/Dear Fager Fan,

You have received an infraction at Horse Racing Forum - PaceAdvantage.Com - Horse Racing Message Board.

Reason: Deleted Post / Thread
-------
And you have been assessed a penalty point.
-------

This infraction is worth 1 point(s) and may result in restricted access until it expires. Serious infractions will never expire.

Original Post:
http://www.paceadvantage.com/forum/s....php?p=2142067
Quote:
Quote:
Originally Posted by HalvOnHorseracing View Post
For your backfitting theory to hold any water you would have to show that the original methodology used to calculate DI was adjusted to fit the historical data. It wasn't. The same methodology was applied to all data.

If your argument is that there was a conspiracy to select chefs-de-race that would produce results to fit the DI cutoff, again you are simply wrong.

You can offer no evidence that data were manipulated to produce a specific outcome. Examining historical data is done all the time in science to bolster a theory.

As for your "standard deviation of the Derby results," are you sure you know what you are talking about? Without going into the math, first you have to calculate the mean of the data set, which for the sake of argument we'll say is the mean of all Derby winners since a respective date. At that point you can calculate the standard deviation, which would represent the distance from the mean of a respective winner.

The worth of the standard deviation calculation in this case is exactly....zero. The measure of success is not the standard deviation from the mean, but a simple DI value below 4.00. If the mean of the winners was, say, 2.50, perhaps that could be tested to see if the farther from 2.50 a horse is, the higher the probability of losing. Perhaps that might merit a standard deviation calculation, as in "horses within 1 standard deviation of the mean won at a higher rate than horses outside that measure," but that is all irrelevant at this point.

That's twice you've used terms I have to believe you are not experientially familiar with. I'd advise you to give up the ghost here and stick to things where you have expertise.
More condescension from Halv. Surprise.
All the best,
Horse Racing Forum - PaceAdvantage.Com - Horse Racing Message Board

[Fager Fan]

Just so people can read what I wrote to receive this ruler slap:

More condescension from Halv. Surprise.

[PaceAdvantage]

Originally Posted by Fager Fan

Just so people can read what I wrote to receive this ruler slap:

More condescension from Halv. Surprise.

Quality post man. I can see why you would be upset at having it pulled.

Look, if you don't want to play by the rules, that's fine. This kind of stuff isn't open to debate.

[dilanesp]

Having aired my statistical arguments, I wanted to share a personal story about Dosage. Because it played a role in my philosophy about betting and gambling and statistics that I don't think about all the time, but it is clearly there.

Dosage was really in its heyday at the same time I really started paying attention to horse race handicapping as an intellectual exercise, in the 1980's when I was a teenager. I started buying Beyer's books, making my own speed figures, watching the replay show at night and taking trip notes, etc.

And I was-- as I am sure some of you have experienced here-- something of a California honk. I thought Trevor Denman was the best race caller, I thought we had the best jockeys (Shoemaker, McCarron, Stevens, Pincay, Delahoussaye), and the best trainers (Lukas, Frankel, Whittingham). And I fell in love with Snow Chief in 1985 as a 2 year old.

I first saw him in a small stakes for Cal breds, which he won easily. Alex Solis seemingly never had to touch the horse. And then he just rose up. He won the biggest 2 year old race in California, the Hollywood Futurity. Then he shipped all over the place picking up purses on the way to the Kentucky Derby-- he won the El Camino Real at Bay Meadows, the Florida Derby at Gulfstream, and the Santa Anita Derby. He was the favorite for the Kentucky Derby.

One of the dosage guys (I think it was Leon Rasmussen) wrote a piece on Derby day in the Racing Form saying that he couldn't win because he had terrible dosage. And he really did-- he was by Reflected Glory, a California sire with sprint breeding, and his damside was full of sprinters as well. He had typical California speedball breeding. On the other hand, I looked at his races in the Santa Anita Derby and the Florida Derby and was pissed at Rasmussen-- "of course he could go 1 1/4 miles! He wasn't slowing down much at the end of these races! He was geared down!".

Of course, Snow Chief finished up the track on Derby day. And of course, Rasmussen crowed about it in the Racing Form. Another victory for dosage. The victor, Ferdinand, was a dual qualifier-- he not only had great dosage (he was classicly bred, by Najinsky from a Double Jay mare, I think) but had been close enough to Snow Chief as a 2 year old to get himself onto the Experimental Handicap.

But here's the thing. I didn't buy it. Snow Chief didn't lose the Derby because of lack of stamina. He lost because of a fast pace. He went out with Groovy, one of the fastest horses on the planet, who later became a top sprinter in New York.

I confidently bet Snow Chief in the Preakness. I mean, I knew. I was naive, of course. There were tons of reasons he could lose the Preakness. For one thing, Ferdinand was a lot better than I thought he was! But I knew....

And of course, Snow Chief didn't get compromised on the pace and won the Preakness at 1 3/16ths, and he did it pretty easily.

But more importantly, he then shipped to Garden State and won the Jersey Derby at 1 1/4 miles. As a 3 year old. Why wasn't this precluded by that precious Dosage theory?

And then as a 4 year old, he ran one of the absolute gutsiest 1 1/4 mile races I have ever seen in my life in the Strub, holding off Ferdinand and Broad Brush when both of them were in top form and when Ferdinand hooked him at the top of the stretch and battled him all the way down.

My point is, Snow Chief could get a mile and a quarter. He always could. And had the pace been slower on Derby day, he certainly could have won the Derby. And any faith I might have had in Dosage as a handicapping system was severely called into question by February 1987, when he won that Strub.

Now I know that this is just an anecdote. And obviously, one example doesn't prove anything. Except-- and this is important-- the advocates of Dosage thought that Snow Chief's loss in the Derby DID prove something, and I know it didn't.

And I've always kept that in the back of my mind, about betting systems. Not only in horse racing, but also in poker, stocks, sports, and other things. Confirmation bias is a powerful drug.

[Fager Fan]

Originally Posted by PaceAdvantage

Quality post man. I can see why you would be upset at having it pulled.

Look, if you don't want to play by the rules, that's fine. This kind of stuff isn't open to debate.

You want to state publicly that you deleted one of my posts and formally scolded me, leaving people to use their imagination about what horrible thing I must've said, then I'm inclined to show them just how horrible it was.

Halv has a really bad habit of talking down to other people here. There have been at least a half-dozen people to comment on it in the past week or two. But yes, it's all me.

[HalvOnHorseracing]

Originally Posted by dilanesp

Halv:

It's not about winners alone.

Take something that does have validity, Beyer speed figures. Not only do they predict their share of winners, but the second place finisher is likely to have a higher Beyer than the 3rd, the 3rd higher than the 4th, etc.

That's what a handicapping tool looks like when it has statistical validity.

In contrast, tons of "systems" produce lots of "winners" over small samples.

The study that dosage proponents never did was to show that Dosage is predictive of better performance, i.e., that the horses with lower dosage tended to finish better (even if they didn't win) than equivalent horses with higher dosage. It would be pretty easy to construct that study, and since it would be studying all Derby horses, not just winners,,the sample size would be far higher. And yet they never did it. As I said, that speaks volumes about what Dosage really was.

(I should add, I don't buy that there's something special about the Derby. An even better study would include all 1 1/4 mile stakes races. The fact that Dosage never worked particularly well in 1 1/4 mile stakes outside the Derby was also telling.)

Perhaps this was my bias. The only time I ever considered dosage was during the Triple Crown season, and the reason was that one of the keys to selecting the Derby winner was to discern which horse early in his 3 year-old season was likely to get the mile and a quarter, especially considering none of the starters had ever negotiated the distance. I'm not sure how much Roman pushed DI beyond the Triple Crown season, but I don't remember having a a lot of discussions beyond the TC. Beyond that, you might have 3 or 4 relevant races on which to base a Derby decision. Certainly races from the 2 YO season had to be given some limited value, but primarily you had races at a mile and a sixteenth or a mile and an eighth from the 3 YO campaign that were what you had to evaluate. In that case, making an evaluation of lineage could prove useful, and even if DI was accidentally correct every year, it still was producing a positive result. There is really no doubt that genetics play a role in a number of things, including athletic ability, and there are only so many ways to take a stab at evaluating that.

Interestingly, Beyer at one point used the same conceptual idea as Roman, that horses below a certain Beyer number were toss-outs, and that there was a minimum number that a horse had to meet to have a chance. The point, and this is where you were right (I don't think I'm putting words in your mouth), was that correlation is not causality. But whether or not there was causality, there were positive results as Gramm and Ziemba concluded. It did correlate.

But as I mentioned in one of the posts about the turf, pedigree is important - once. Once the horse has successfully negotiated the mile and a quarter, from that point I'm fine assuming the horse can get the mile and a quarter regardless of breeding.

The irony here is that the discussion is now moot. Roman has given up on DI, breeding has made it more obsolete every year, and the dual qualifier concept is a nostalgic concept. Next year the discussion may be, I wish I had a dosage number to consider. Or not.

[HalvOnHorseracing]

Originally Posted by Fager Fan

You want to state publicly that you deleted one of my posts and formally scolded me, leaving people to use their imagination about what horrible thing I must've said, then I'm inclined to show them just how horrible it was.

Halv has a really bad habit of talking down to other people here. There have been at least a half-dozen people to comment on it in the past week or two. But yes, it's all me.

I'll offer my best advice to you from experience. If you think someone is a blowhard, and you are interested in shutting them up, make a better argument.

[dilanesp]

Halv:

Don't give up! Pedigree IS important, even now.

A couple of years ago, Commissioner finished second in the Belmont. His record didn't suggest a second place finish was likely. But check out his breeding.... (His stud fee at Winstar is less than $8,000, I think. If I had a mare, I'd breed to him....)

The reality is-- and you adverted to it earlier in this thread-- that there are handicapping factors that have importance but can't be quantified. We all know, for instance, that watching previous races is one of them. There's no specific number that you can enter into a database for a horse who gets strangled by her rider or pinched back on the turn or caught in a speed duel. You just have to make a visual assessment of it and factor it in.

You definitely should take into account breeding in handicapping a horse on the stretch-out. And that's broader than the Kentucky Derby. I look at breeding when horses are going around 2 turns for the first time, and sometimes even when horses are stretching from 6 to 7 furlongs. I look at it when a turf sprinter is trying a turf route for the first time.

The problem with Dosage was NOT that they were identifying a factor that wasn't relevant. It is relevant, even now, whether there is stamina influence in the pedigree. You just can't be dogmatic about it. Some horses (like Snow Chief ) outrun their pedigrees, or, as you say, catch Derbies full of horses who are just as poorly bred for the distance as they are. Dosage's failure was that it was a dogma, putting a number on something that wasn't really quantifiable.

[HalvOnHorseracing]

Originally Posted by dilanesp

Halv:

Don't give up! Pedigree IS important, even now.

A couple of years ago, Commissioner finished second in the Belmont. His record didn't suggest a second place finish was likely. But check out his breeding.... (His stud fee at Winstar is less than $8,000, I think. If I had a mare, I'd breed to him....)

The reality is-- and you adverted to it earlier in this thread-- that there are handicapping factors that have importance but can't be quantified. We all know, for instance, that watching previous races is one of them. There's no specific number that you can enter into a database for a horse who gets strangled by her rider or pinched back on the turn or caught in a speed duel. You just have to make a visual assessment of it and factor it in.

You definitely should take into account breeding in handicapping a horse on the stretch-out. And that's broader than the Kentucky Derby. I look at breeding when horses are going around 2 turns for the first time, and sometimes even when horses are stretching from 6 to 7 furlongs. I look at it when a turf sprinter is trying a turf route for the first time.

The problem with Dosage was NOT that they were identifying a factor that wasn't relevant. It is relevant, even now, whether there is stamina influence in the pedigree. You just can't be dogmatic about it. Some horses (like Snow Chief ) outrun their pedigrees, or, as you say, catch Derbies full of horses who are just as poorly bred for the distance as they are. Dosage's failure was that it was a dogma, putting a number on something that wasn't really quantifiable.

Some people know I am good friends with Doug O'Neill. I very respectfully questioned whether Nyquist could get the mile and a quarter. I thought his future was as a miler. O'Neill told me, no problem. He's a mile and a quarter horse.

In the end, I thought his competition was pretty limited and I wound up picking him on top, but I had nagging doubts about whether he would hit that last eighth wall right up until he crossed the finish line.

I was clearly wrong about whether he was bred for the mile and a quarter if you go by results. Of course, if Roman was still toiling, he probably would have reclassified Uncle Mo as Intermediate or Classic and he'd have looked dominant on DI.

I'll still consider breeding. I'm not sure precisely how.

[PaceAdvantage]

Originally Posted by Fager Fan

You want to state publicly that you deleted one of my posts and formally scolded me, leaving people to use their imagination about what horrible thing I must've said, then I'm inclined to show them just how horrible it was.

Halv has a really bad habit of talking down to other people here. There have been at least a half-dozen people to comment on it in the past week or two. But yes, it's all me.

Fair enough. I probably shouldn't have posted anything publicly...I kind of assumed you would post about it publicly or question why your post was deleted, so I decided to beat you to it.

It's a learning process.

[Fager Fan]

Originally Posted by PaceAdvantage

Fair enough. I probably shouldn't have posted anything publicly...I kind of assumed you would post about it publicly or question why your post was deleted, so I decided to beat you to it.

It's a learning process.

I'd leave just the except I understand that could get me another demerit from CJ.

[Waquoit]

Dosage was baloney. It worked until it didn't. It only worked for me when Real Quiet won at 6-1.

[HalvOnHorseracing]

Originally Posted by Waquoit

Dosage was baloney. It worked until it didn't. It only worked for me when Real Quiet won at 6-1.

You're a little late to the party. We've already beaten this one to death.

[PaceAdvantage]

Originally Posted by Fager Fan

I'd leave just the except I understand that could get me another demerit from CJ.

There is no category in the points system for an "emoticon only" posts...so you're safe. LOL

Posts deleted for minor reasons such as emoticon-only or because it's off-topic to the thread should not be assessed a penalty.

[WoxFan]

I was interested in the Performance Figures that he had worked on and was posting. Anything that is out of the box and takes a different look at measuring the performance interests me.
In general, anytime someone walks away from this game that had a lot of passion for it at one point is a shame.