I'm hoping for some guidance on effectively using the ratings generated by PPro Gold V1. Can anyone tell me from backtesting how close to accurate the rated prices are? eg do horses rated at evens have close to a 50% strike rate? do 2-1 rated horses win 1/3rd of their starts? If not, what adjustments do I need to make? eg does 10 start form produce a better result than 3 starts? does altering the jockey ratings help? It seems to me that if the ratings aren't close to actual probabilities then they are not very useful to use in looking for genuine value bets.

Daz
I think you ask for predictive perfection.
But these factors need to be remembered:
- predicting on a total range of different races with very diffent amount of form and class, one would expect a computer generated output even as comprehensive as Price Predictor Pro to vary considerably in its assessment strength. My experience is that the better the class and form of the race the closer the assessment is to outcomes. I expect wider variations of results of poor quality races with the high degree of inconsistency of horses in those sort of races. Same with low form races, very open races maidens etc.
One of the great strengths of P.P.P. is its system selector. Narrowing down to the better form races its prediction is closer.
It is a very good guide and tool, but to expect perfection of prediction is not possible anywhere I don't think.
But with it, and using a bit of nous, you are a long way ahead of many others.
In the last three Melbourne Cups the winner was in the top three predicted by P.P.P ... and that is in a notoriously open race.
I personally don't go the overlay way, but use its top assessments to get to the core of a race in my betting.
It is also a marvellous tool also to learn heaps about what are the priorities in assessing races. I am still learning as my data base becomes respectable.

Regards

Rod

Good question...most winners pay Less than PP std rated $( I hope ? Iv'e got that right Neale)...so it comes back to how you use the information...Many only go for overlays...I'm not convinced...as I feel you miss to many good opportunities..But I like volume..which is an issue or stratergy in itsself
cheers Zach

__________________
Regards Zach

Thanks for the replies. I didn't mean to be critical of the program, I'm just trying to learn to use it better than I do at the moment. My database isn't big enough to backtest the accuracy of the ratings and I was hoping for some guidance, which Rod has given. Logic tells me that if the ratings were close to accurate (but not necessarily perfect), then using an appropriate buffer for inaccuracies I could stick to betting overlays and come out in front. Clearly this approach would be hopeless if the ratings were not close to accurate. It seems that my approach may only be suitable for certain higher class races. I'd still be very interested in any stats on the accuracy of the ratings in different types of races. Thanks again.

If you could change the market setting to 155% this will reflect closer to what the newspaper percentages would be & feel you would pull more winners betting the overs .

This will minimise the number of the unders that get up.
Its best to target races where the prepost fav has a good chance of falling over so as to attain better value on the other selections if & when they get up.

__________________
Cheers.