I disagree, there's far too many combinations of filters for the marketplace to be overwhelmed and hence the horse to be well under the odds as a result. Also, punters, being mugs have made the same mistakes since the year dot, and will continue to do so. CP statement that it's best attribute is highlighting what doesn't work, is exactly what I've found too.

I have found a nice few systems, with few bets, that are not all purely mechanical because 1) they give a horse a run or two after being selected (and this incorporates one of Don Scott's major axioms (as I've interpreted it)) and 2) I look for horses from NZ, o'seas (& up to 3 preps here) with a few filters to get rid of those targetting Melb Cup for instance, and interstate horses.

So, the database can be used in conjunction with any method of selection.