you know over many years i have been where this argument leads e.g. all sorts of formulas and hard work and still the good prices elude you for various reasons.
the flaws i have found in trying to formularise racing is that most data is obsolete although it is mainly the only thing you have to go on.
e.g. good horses like sports and others will win at big odds. they have good group 1 form won lots of money good percentages but generally hard to catch.
flavour is another. but thats beside the point. i have a program written in perl that searches cyberhorse for a particular strength in the fields that i look for.
it is hard to win much mainly due to my lack of discipline and my need to try to ba ck too many winners. is the picture becoming clearer. after many years of formulae, speed, times, weights i concluded there was only one factor that really delivered consistent winners. you can save yourself all that study. but to give the game away it is not what i want to do but i'll give a clue.
makybe diva , british buska hardrada choisir media puzzle ra sun all had the common factor as did many of the horses that lost on saturday.
its hard work isn't it.