My thoughts are that things can be a little over analysed and complicated by punters.

Keep detailed records.

Is it possibly as simple as SR [in % terms] x Ave Ret = [Over 100 good] [Under 100 bad].

Punters introduce a price component when it's Under 100. That is the selection method isn't profitable. But that becomes rather dangerous, because as mentioned, you invariably pass up winners. Now backing less losers than more winners is always my aim [to a point] but how do I really know which horses to pass up? Do I do it on mathematics or some other means, which is heading down the system punters route of being forced to introduce more rules to try and eeek out a profit.

When all this is saying is you're selection method is no good...

I tend to think punters would be well advised to look at selections that are further out in the market [can you argue that introducing a pricing component is the same thing? maybe?].

I feel that plenty of market favourites and jockey/trainer favourites are under the odds. Some of the less fancied runners can also vary more widely in markets providing value [in theory].

Plenty of ways to skin a cat and I'm sure others will have different experiences but that's my [rambling] thoughts :-)

Edit: posted before seeing your last post Stix but agree with some of your throughts about Ave div and SR