I haven't had any success in isolating the Winning Races from the Losing ones in my studies.
I figured quality of runners and amount of data available were good criteria, but I been trying to deny I was wrong, but my Bank Account over the years testifies that it ain't true.