Database testing of jockeys

[hdcper]

Hi Everyone,

Today at the track, several handicappers and myself began discussing jockeys and how we would measure their overall ability. One handicapper felt that maybe the best way to measure their ability was to determine their win percentage when riding post time favorites and use the stay fast 33% win rate of favorites that has existed year in and year out (in other words a control for comparison purposes).

Anyway I thought one of this elite group with a database might be willing to run a test and post the results. If possible, also providing the ROI on the favorites for each jockey would be helpful.

Any other ideas of evaluating the jockey would also be helpful,

Hdcper

[danzig]

The really good jockeys, the honest ones who have been around, will all tell you that they are the passenger and the best of them INFLUENCE the horse the least. Once you have done that, the horse runs the way it can but rarely does. When Bailey won the B.C. Classic on Arcagnes, it was because, as he said it, "I didn't impose my will on the horse's natural running.

In a nutshell.

Tim Yatcak

[Rick]

I think the best way to measure jockeys would be by how much better or worse the speed rating of horses are when they ride after a switch from another jockey. The idea of using favorite win percentages was mentioned at least as early as 1976 by Gordon Jones, so it's not exactly a new idea. But it may have some merit since it's not readily available to the public. You could also use ROI and maybe adjust for number of wins or mounts.

[David McKenzie]

McCarron prolly said it best, "All jockeys are detriments to horses."

There are exceptions to every rule. Ever wonder how Russell Baze manages to split the eye of a needle and "push" all those horses home? He's g-o-o-d, and so is his agent.

And while most jockeys can't be said to win races, they can certainly lose them, especially on the grass.

My two cents...

[cj]

Originally posted by hdcper
Hi Everyone,
One handicapper felt that maybe the best way to measure their ability was to determine their win percentage when riding post time favorites and use the stay fast 33% win rate of favorites that has existed year in and year out (in other words a control for comparison purposes).


How would you rate the jocks who rarely if ever ride a favorite?

CJ

[Tom]

One way to decide how good a ride was might be to compare it to the odds. IF a horse finishes 3rd and is the fifth choice, it was a +2 place better ride the odds suggested? A favorite runs 3rd, it is a -2 ride?
This would be something you could get out a database fairly easily, I think. I read about this somewhere years ago, I tink it might have been Racing Weekly, or whatever it was called - the one that used to have Pandy's variants each week.

[cj]

Tom,

Only problem with using the odds is the jockey factor is built into them. In some cases (Baze, Day, Bailey, Bravo) it is the biggest factor. One of these riders would almost always get a - due to choice of mounts and betting action. I've been thinking about this for a better part of the day and keep coming up empty, so I hope the discussion continues.

CJ

[BillW]

Another issue would be trainer bias ... top jock gets the well meant horse.

[andicap]

Also jocks like trainers have strengths and weaknesses. You'd have to use subcategories and not just judge jockeys broadly. A speed rating method is flawed because there's no control. There are so many other factors that affect how a horse runs.

Besides sometimes the best thing about a jockey is his agent. When Mike Smith was winning everything in NY a while back, many people thought it was because his agent Steve Adika was particularly adept at putting him on the best horses.

[hdcper]

Tom,

The publication you were talking about is Racing Action if my memory serves me right. I really enjoyed that weekly publication!!

Enjoying the feedback on this topic, but still hoping a database guru will share some of there ideas and results.

Hdcper

[tdthomas]

One thing I would like to know is, How often does each jockey get disqualified. Do bad jockeys get disqualified more than good ones? Or are good ones more aggressive, causing them to be disqualified more?

[Tom]

There was a weekly newsletter years ago called Woodside something or other, and it was from Tom Ainlse and Woodside Associates. They did a study on how jockeys did in photo finishes.
They only got rated if they were involved in a photo and then the results were tabulated a lot like give-aways-take-aways in football, so a jock might be a 4-6 or a 2-12.

[GameTheory]

An aside about Woodside:

Woodside sells the Henry Kuck ratings:

http://www.woodsideassociates.com/


They don't seem to offer anything else at the moment.

[ranchwest]

There are a lot of factors for a jockey.

I've had a theory for a long time, but I've never tried to prove it. I think a jockey needs more strength in a sprint and more rhythm in a route. If that is the case, you would have to rate jockeys separately. There also appears to be a big difference between dirt and turf.

I've also seen jockeys like Ronald Ardoin who I've long felt is only interested in the win position. If he's not riding for a trainer with whom he has a relationship, he seems to give up if he feels he can't win.

Then there's guys like Day. He knows how to navigate through traffic and he deserves his nickname, "Wait All", which helps him bring in a lot of winners.

Then there's guys like Corey Lanerie. Have you seen his wife? Not much doubt about why he wins. :)

I just don't see how you can rate a jockey numerically and have it truly disclose his competence.

I look at win percentage in 30 days just to see if the jockey has been hot.

[Dave Schwartz]

Some years ago I built a database of "photo finishes" for each jockey in our database. What I found was interesting:

1. Because of blanket finishes, 43% was the average win percentage.

2. There was a significant difference in some jockey's numbers from sprint-to-route and turf-to-dirt.

For example, Pincay was a world beater in sprints, with a percentage in the low 60s, while he was only average in route races. This came at a time when Pincay was struggling with his weight badly and the rumors said he was also struggling with his endurance.

Another example: The first time I did the study was by hand and it was the year that Desormeaux arrived in SoCal. He was terrible! And it showed in his photos, winning only 25%! Two years later I re-did the study and Desormeaux was now in the high 50s (and had become a leading rider).

Example 3: Pat Valenzuela was also a 25% rider that first study... at a time just before a major suspension. In the second study, he was back and clean and his scores went up to the 60% range.


I think the photo-finish ideas has much merit.


Regards,
Dave Schwartz

[turfspec]

My idea of a competent rider is one who gives his mount every chance to win. That is, he keeps them out of trouble, saves ground when possible or finds the best part of the track and places them where they need to be to give their best. With that in mind, I go through the charts and PPs and consider every mount for a rider going off @ odds of 5-1 or less. Then I look for trouble - bad break (if the horse isn't prone to bad breaks), clipped heels, in tight, lost whip, unnecessarily wide (judgement call), failed to maintain path & bumped among others. Much of this is subjective but I strive for objectivity. With enough races it should tend to even out. The average for top 10 riders at major tracks is about 14% unexcused troubled trips. For example; a study I did at Saratoga showed J. Bailey @ 11% (he's usually the best @ 9%-12%), J. Velasquez & E. Prado avg. @ 14%, J. Chavez slightly above the avg. @16% and P. Day nearly 3x that of Bailey at an astounding 32% over a 2 mon. period - far and away the worst among top riders.

Rob

[andicap]

Turfspec,
I don't understand the reason behind your research, fascinating as it is. P. Day gets into more trouble than most riders, but still wins more than most. So what's the point??:confused:

[ridersup]

Dave:

I'll bet Jerry Bailey would get good marks in your testing. He seems to get the last bit of energy a horse has left at the wire.

At my track T. D. Houghton has the same ability and seems to be able to hold these lower class horses together while no one else can. Consequently most of the trainers in the area flock to him.

[delayjf]

Personally, I look a Jockeys as more an indication of trainer intent. I'm sure that some are better than others. I know in SoCal, I never bet a turf horse ridden by Martin Pedroza, he's a speed rider with no sence of pace. I used to think the same about Ken Desormaux (excuse the spelling), but he has gotten much better rating a horse and coming from behind, on dirt and turf.

On another Jockey note, I think too many of them think that they need to get to the rail at all cost. Yes, running wide does hurt a horses chances, but so does not being able to begin your move until it is too late. One thing I've noticed about Jerry Bailey is he almost always puts his horse in position to run his race, he seems to rarely get into trouble. I don't know how many times I've cursed Jockeys who plant themselves on the rail behind two echelons of horses, who do they think they are, Moses? On the plus side these riding "Stiffs" often comeback no worse for the wear.

[turfspec]

Andicap,

Not sure how to answer you. As I said, I prefer riders who stay out of trouble giving their mounts every chance to win (First do no harm). Also prefer riders whose wins exceed places & places exceed shows. Most riders who have a higher than avg. trouble index also have a lower than avg. win%, so you're right, Mr. Day does present something of a dichotomy in that respect when you view his record in total - but jockeys like trainers or handicappers have strengths and weaknesses. They can vary by distance, surface, circuit, track or season. The vast majority of Mr. Day's wins have come in Kentucky, especially CD, where he dominates like Bailey in NY and Baze in No. Cal. (Despite his rep for waiting, @CD most of his wins come on or near the front.) If you expect him to ride to the same percentages @ Sar as he does @ CD you will probably be disappointed.

At the just concluded CD meet Day had 245 mounts with 72 wins, 42 pla & 33 shw. (note the descending placings) a 29% win rate. Just finished looking at the top 18 riders at the 2001 Sar meet, separating rider performance on turf vs dirt. Day overall last season @ Sar was 156/ 23 - 33 - 27 a 12% win rate. Significant difference? Broken down by surface: Turf 64/ 10 - 5 - 12 19% win rate. (Believe 3 of the wins were wire-wire. Tough to get in trouble in front) Dirt 122/ 13 - 28 - 15 11% win. If he weren't so prone to trouble would more of those places have resulted in wins? You'll have to watch the races, read the charts and decide for yourself.

Rob

[andicap]

Only problem is that many riders don't get good mounts below 5-1 so are hard to rate. They might be good riders but since the top mounts go to Baileys of the world, the competent, but underbet jocks don't get rated under your system.
I'd like to find a way to evaluate the underbet jocks, the ones who pay big balloons because bettors are focusing on the Days and Baileys of the world and ignoring the Castallanos.

[ranchwest]

I like a jockey who will give a big try without getting downright excessive. If he gets into trouble sometimes, but wins a lot, that's OK by me, at least he's trying.

One of the reasons stats are so close at SAR is that the competition is tough and it is difficult for jockeys' agents to get their jockeys very many sure winners. It should not reflect poorly on a jockey if he is, say, in the top 5 at CD and SAR, but has a lower win percentage at SAR.

In using statistics, it is important to understand whether you can reach a sound conclusion, that there indeed is a correlation. Sometimes there are additional factors and sometimes backfitting the data simply discloses a coincidence.