That could be a good idea but from a statistical point of view you would be better off selecting 50 horses who have had between 4-9 runs in as they represent less of the runners but more of the winners...36.2% as opposed to 1-3 which win 28.5% of all races.