ubetido,
yes I have been trialling a method similar to this but as you say there are variables which need to be factored in. Prizemoney alone is not enough, you must combine it with weight carried, consistency, recent form and lengths beaten (very important).

There are probably very different viewpoints from everyone on the relative importance of each factor, that is why 10 people might have the same criteria, but come up with 10 different selections.
As to improving horses, generally they will not beat a class horse which is in form. I say generally statistically speaking. Improving horses usually go through the grades very quickly and as such their prizemoney results from wins or very close placings if they're really class horses. So a class improving horse would probably have better recent prizemoney than say a poor performing Bomber Bill etc. That's why I say recent form is important as is recent prizemoney. Quite often very good horses will come back from a spell well below their previous potential (Calaway Gal or Quays for example).
It is important to recognise that there will be exceptions to just about every rule you can think of, but you have to go with the weight of statistics.
An improving horse will beat a class horse below it's best form though and usually at decent odds.
There will be those that debate my views and those that agree, that's just the variables of racing. You have to come up with something you're comfortable with and stick to it if it's profitable longterm - not give it away on one bad week.
Having said that, you must constantly look for ways to improve the longterm result. Trends and patterns change quite frequently, which is why a lot of systems based on past results fail.

[ This Message was edited by: Equine Investor on 2002-11-13 01:42 ]