Jockey Intent
 

Jockey Intent has to be figured into the equqtion somehow, but I dont how. Ill try to express my point
Yesterday 1st race @ SA #7 Royal Opera House ridden by Kent D
Now the horse was claimed off of Bob Hess who Kent reguarly rides for. The horse was claimed by Alfredo Marquez(who) and Kent and I guess his agent ask to ride back. The horse won and paid $11.80, an underlay from ML, pointed out by Kurt Hoover(who picked the horse for some of these reasons), and an overlay on a KD ridden horse, probably because of the no name trainer.The horse finished 2nd last out , same condition, @ 3-1, with KD.

I believe both BRIS and DRF sell downloadable data for data base work.

Both sell csv files of historical data, which makes it easy to do the db work. Having gone with DRF myself, I'd go with BRIS if I were starting over. BRIS provides much more data, including workouts which is huge. DRF csv files don't have workout info and purchasing separately is very pricey. I think BRIS yearly plan is actually cheaper as well...

I think the ability of the jockey should be factored into the overall rating for each horse, those horses converted to an odds line, and the odds line determining betting. At no point do I consider how much the jockey gets bet by the public. It will be factored into my betting line already.

Simple example, lets say you use two ratings, horse and jockey, and count the jockey for 20%. Horse A is rated 100 and Horse B is rated 100. Horse A has a jockey that rates 100 on ability and horse B has a jockey that rates a 50. Overall ratings are Horse A is 100 and Horse B is 90. Convert those to an odds line and see how it plays out. I now the public will bet Horse A more because of the jockey but I also know that it is warranted. You need a premium to bet on Horse B in my opinion.

My point is that at no point do I consider if the jockeys are overbet or underbet normally by the crowd. I think that will all come out in the wash with a good odds line and comparing it to the public odds. But I definitely factor in the ability of the jockey.

One thing I would caution is that you need to factor in the jockey in past races as well. If Horses A and B have been continually ridden by the same jockey in previous races, make sure those 100 ratings reflect that as well.

that's what i was trying to say!
you just said it better than i could have.

Great thread
 

A buddy of mine contacted me and thought I was posting in this thread. Several years ago, I developed a jockey rating system based on similar methodology as discussed here. I feel the ratings are good, but there are excellent ideas here and a better system underway. Years ago, Turfday.com had terrific jockey ratings that I love to used. (It also had excellent sire and trainer ratings too). After Turfday went out of business, I searched everywhere for jockey ratings that could add some betting value to my game and finally came up with my own. I'm looking forward to seeing what's created here -- it would be great to have an automated option with more tracks available and less manual work needed.

When I first started playing this game I used to pay close attention to jockey switches, jockey trainer combos, and red hot riders getting a lot of live mounts. Then somewhere along the line I started paying very little attention to jockey at all. I guess I didn't feel like it was adding much value to my opinions.

As an owner with a piece of a few horses, I'm paying more attention again. I've had the opportunity to observe a few jockeys. I know who has listened to instructions, who is riding hard for minor awards, who is being more or less aggressive than I want etc... Now I find myself forming opinions again. After a race I'll immediately start mumbling to myself how my horse would do better with so and so as a rider.

I do not imply that the jockey should not be factored in the rating of each horse; instead what I am trying to say here, is that this factoring should not depend it his absolute ability but how this is been conceived by the betting crowd.

I am also saying that exactly the same principle applies to any other applicable handicapping factor that affects the betting patterns of the crowd.

Take as an example a horse that is offered at 4-1 odds; this means that the crowd believe that its winning chances are 0.2 or 20%. Whether this horse represents an overlay, underlay or a neutral betting proposition depends on how its handicapping factors are perceived by the betting public and not by the it’s (always unknown) absolute winning probability. The jockey’s ability is already part of the 4-1 odds since the crowd is well aware of its significance it the outcome of the race. The real challenge is not to quantify the ability of the jockey but to detect potential estimation errors committed by the crowd.

To keep our example relative to the topic of the thread, let’s assume that this starter is ridden by the absolutely best jockey that can be found in the circuit. The ability of the jockey alone, is not enough to derive any kind of useful handicapping opinion. It can very well be the case that the top ridder represents a very valid reason to bet against him; this happens when his superiority is so obvious that the crowd is mislead to the point of overbetting him while simultaneously creating value in some of the other starters. The reversed situation can very well occur in the case of the worst jockey who might be tremendously underbet converting his mount to an overlay.

Exactly the same behaviour applies to any other handicapping factor; think of females against males or a claimer who is trying stakes company for first time. In either case, it is well known that the winning chances of such a starter are significantly diminished but this does not necessary imply a bad bet; in many cases quite the opposite is true! It all depends on how the crowd will perceive each situation and to what degree it the bets will be optimally distributed or not.

Following this way of thinking, it is easy to conclude that what matters when it comes to the gambling aspect of the game, is not the raw ability of the jockey but the degree of the potential crowd mistake in its evaluation.

There are 2 common approaches to finding value.

1. Figure out how important each factor is in predicting the winner and then combine them into a model for making an odds line to determine your value plays.

2. Try to find factors/angles that the public often under/over estimates and use filters and combinations of them to find value plays

If you use approach #1, then you are probably interested in a rating that measures the skills of the jockey that you will combine with similar data on the horse, trainer, etc...

If you use approach #2, then you are interested in which jockeys are under/over bet in certain situations.

this is a great post. it's like kitten's joy offspring. they like turf but nowhere near as much as the betting public thinks they do.

Have to say 'Thank You' for doing the work, and sharing it

Surprised to see Jose Cruz looking great, and Saez looking terrible.