Speed Figure Odds Line Software

[sonny]

Does anyone know of any software (either publicly available or in private hands) that will generate an odds line based upon recent speed figures (Thorograph or Ragozin) of the horses in a given race?

[hurrikane]

try the mlo in the daily racing form.

[kitts]

Check out "Figline" at:
http://www.netcapper.com

[GR1@HTR]

Don't know anythign about this one: http://www.thorograph.com/truline/index.html

I tried fig line on a limited basis and found it to be a little strange in the oddsline it create, just my opinion though....


but the best by far is to create your own odds line after reading steve fierros 4 quarters of horse racing.

[Speed Figure]

Fig Line is very strange you could never bet a horse at the odds it gives because they just don't go off that high.

[Holy Bull]

sonny:

I have a screen in my s/w where you can enter or import a projected Thorograph figure for each horse and it will create a pretty strong odds line. Is this what you are looking for?

[sonny]

That's the type of thing I was talking about. Can you please tell me more about how it works?

Sonny

[Holy Bull]

Sonny:

How it works is you feed it the projected figure you think the horse is going to run on each horse, and it will run a simulation on the race and spit back a win odds line and a place odds line. It will also suggest bets based on this odds line vs. the current live odds if you want it to.

Drop me an email at holybull78@hotmail.com if you are interested.

[Lefty]

Ray, why would a mathematical calculation be best done by a human. Once the formula is in place can't a computer do it faster, more accurately?

[HTRFGuy]

I spent a year looking at handicapping factors and testing before I was satisfied with my computer generated odds line that I incorporated into my HTRF handicapping software.

The final odds lines is programmed to examine 48 individual factors for each entry in the race. Many are based upon pace matchups for the current race which, in turn, are based upon your selections of past performance running lines for handicapping.

Nevertheless, I get spurious results especially with lightly raced horses.

E-mail me privately if you want to know more about the factors that go into the odds line.

[ranchwest]

HTRFGuy,

Sounds like you are using 44 too many factors. :)

[Holy Bull]

The handicapping process and creating a odds line are two seperate parts of the equation. If you are trying to take 50 factors and generate an oddsline, its probably not going to come out well. If you can isolate the two processes, to where you can handicap a race down to just a few or even 1 rating that can meaningfully and consistantly distinguish quality, create a strong odds line becomes a purely numerical exercise (although one that few that I've seen have done well).

[Holy Bull]

Not being able to properly train a computer to follow your own methods does not mean that it can't be done, nor does it make them art.

[GameTheory]

There is nothing wrong with using lots of factors (or variables -- you might have 10 variables that are part of a "pace factor") if you're using a computer to generate an oddsline. It works just fine if you do it right....

[HTRFGuy]

Originally posted by GameTheory
There is nothing wrong with using lots of factors (or variables -- you might have 10 variables that are part of a "pace factor") if you're using a computer to generate an oddsline. It works just fine if you do it right....

The point of my post was merely to indicate my belief that making a computer generated odds line in all liklihood will involve more then merely one or two factors.

I have no quarrel with a human odds line, my objective was to make an odds line for use by users who cannot, or elect not to, make an odds line of their own.

The bottom line is how well your odds line holds up and this can only be ascertained by testing your line over perhaps 1,000 races or more. If the horses that you make 2:1 win 33% of their races then you have a good line. But you also need to test the 10:1, 20:1 and 50:1 to assure yourself that they win ~ 9%, 5% and 2% of their races. It gets dicey when testing the 3:2, 5:2, etc. horses as well. Only when your line gives your consistent results in the real world ranging from odds of 1:2 to 100:1 can you be satisfied.

One last point about the 48 factors, only a small number of these factors will apply to any given entry, perhaps 10 or so. Also, each factor has it's own weight (importance) based upon the attributes of winning horses over many races. Some attributes are positive and others are negative.