I think in post from Shaun he mentioned using recent przemoney averages rather than career averages, logic says that it should be a more accurate guide. Does anybody have a database with the prizemoney each start that could compare the two such as strike rates and profit/losson turnover using the last 12 months against career average?

Secondly, could anybody tell me a website with free form history that gives the prizemoney earned each race or at least what each race was worth? I use ozeform but they dont have this info on their horse search site.

Thanks from the doctor.

Doc,

I have found that ranking horses by Career API and then by last three starts API is very powerful.

The last three starts is the most indicative, career can be based on one or two good 2yo starts.

You can get prizemoney earned per run and the actual value of the race from Cyberhorse html zip files.

dr.ron,
it does seem obvious but you'll be disappointed with using last 12 months and/or recent x no. starts.
For about a fair while I was calculating an API over last 10 starts, and overall didn't find an 'edge' over career API.
I then used distance as a factor. So if a horse had won a 2400m race then was resuming over 1200m I didn't include it in the calculated API. The odd race seemed promising but overall again vs career API was disappointing. Even allowing for some 2yo days and high prized restriced races you'll find time spent on API other than career API won't be worth the effort. Of course please find out yourself, and give Chrome Prince's theory a test run. Maybe a Baghwan or Becareful might have some stats to help out.

Hi anchor,

I found in my analysis that the last three starts was the most indicative and effective.

Going to more than three or less than three just didn't produce the same good results.

It was also more effective if:

Last Three Starts API was greater than Career API (The horse being in the best form of it's career).

To give you an example favourites that are ranked third by last three starts API and were beaten by less than 1 length last start continue to make a level stakes profit on Saturdays over three and a half years of data with no other filters :wink:

Also this has to be used in conjunction with other factors not on it's own to succeed.



_________________
"The man who can correctly assess Class can have all the money he wants and he only needs a dollar to start". - Pittsburg Phil

[ This Message was edited by: Chrome Prince on 2004-03-02 06:03 ]

Thanks for the input guys, I'll keep plugging away.