Vortech, I'm glad you're posting. I like a debate on theory. I don't see anything wrong with "backfitting", whatever it's called, massaging filters to suit an outcome ..... because if it's over a long enough period then the correct / decent / profitable filters will eventually be isolated. I actually / niavely believe that you can have the most stooopid of filters in there for starters, and time will eliminate them as having any influence in finding a decent S/R / healthy div. I've tested systems that work in Vic and don't work in NSW, this is a common occurance, and wierd too. I've concluded that the handicapping class systems between the states contributes to the vast difference in results. Who knows ?? ..... I know I don't on this issue. Any thoughts Vortech ??

A couple of my good systems, (the ones that don't win 'coz I'm confused about POT & ROI, cannot win, won't hold up, have too few selections etc) are based on lightly raced horses. One selection method I have which isn't a mechanical system, and which I've posted on here, is based around lightly raced horses. I've read a lot of Don Scott in recent times, and one of his maxims (my choice of words) is to bet heavily on lightly raced horses, obviously other conditions being met, but he did say just that. I find this encouraging, as it has been after the development of my selection methods that I've encountered his writings. Sort of encourages me enough to know I'm in the right territory having confirmed that DS thought the same 30 years ago, adds a bit of longetivity to the concept. Also a few posts on here regarding the number of selections (sampling etc) have been encouraging too, not that I needed any confirmation in this area.