Joe Zambuto program

[redeye007]

I've settled on the best way I can come up with to self tune. for a new weighting criteria I first segment the races by surface , condition, distance, purse and race type. for sprints I use <=1540. for routes I use >=1760. I'll keep 7 1/2f races in their own category. next I check q-check. click q-check several times. now I check f-tune and click until the new win percentages don't change. I'll note the win percentages. now I'll set all the factors to -1. I 'll click f-tune several times until the win percentages don't change. I'll note the win percentage. now I'll change all the -1's to 0's and run f-tune again and note the win percentages. the method with the highest win percentages of the 3 will be the one I name and save to use to handicap this specific type of race. for existing criterias I just load , tune. and save. Anyway, this seems to be the best way to create the most precise weighting criterias.

[bcgreg]

Are you not afraid of backfitting the data or are you testing against 2 different sets of data?

Regards,
bcgreg

[chrisl]

Hey Red: are you using pre-set on your Q check..Chris

[redeye007]

Originally Posted by chrisl

Hey Red: are you using pre-set on your Q check..Chris

yes and I self tune the 3 ways I described and save the one with the best win percentages. the results have been quite excellent with a minmum of at least 40 race cards and results. I'm also using purse ranges to acquire my minimim of 10 races of my target race type.

[redeye007]

Originally Posted by bcgreg

Are you not afraid of backfitting the data or are you testing against 2 different sets of data?

Regards,
bcgreg

I'm only testing against 1 dataset. I want at least 10 races to create my models. more is better. overfitting can occur using too few races to create a model.

[Partsnut]

The Power Of The Zambuto Handicapper

I posted My selections to my web page 30 minutes before post time.
Here's the results
1- 44% overlayed winning races
2- $1.0917 return for each dollar bet.
3- 6 winners in the top 2 choices (67%)
4- 4 winning overlaid bets out of 12 bets (33%)

[redeye007]

that last pick 3 was a decent $283 for $16 wager 4-1-9.

[redeye007]

Here's another tip for users that I'm seeing positive results with. always look at the pace report in conjunction with your handicap picks report. the pace report displays horse paceline fractions in feet per second at exact distances. If your top contenders on your picks report are beaten in the pace report by another horse at the exact distance then that horse is a must use in any betting plan. they are usually not the favorites

[Partsnut]

Some More Good Results Using The Zambuto Software

[redeye007]

the program made these supers look easy today at Delaware. the top 4 boxed actually hit in the 7th and I used filters in the 8th to hit this biggie on my second ticket.

http://www.paceadvantage.com/forum/s...ad.php?t=75469

[Partsnut]

Zambuto Pool Analysis, Betting Trends Tool

This tool allows the user to run the tote board in real-time and has a segment of his program that runs with the tote board and allows the user to see the actual betting trends simultaneously.
You can view a video of how it works at the following link:

http://www.box.net/shared/r3mbcaxamg

[chrisl]

Hey Parts: I have the new upgrade and have not seen the trend tool. I downloaded the video link you posted but no sound. Any help Chris

[Partsnut]

Chrisl:

Hey Parts: I have the new upgrade and have not seen the trend tool. I downloaded the video link you posted but no sound. Any help Chris

Chris, I'm not much on making videos but you have to access that module which is listed torwards the bottom of advanced features.
There are 3 steps you have to follow and you can get them in the help files.
If you send me an email I'd be happy to send the help file to you.

[Partsnut]

Chris, I just emailed the help files to you.
You should not have any problem.

[chrisl]

thanks Parts