ye guys of no faith. Niche markets it's where it is at, OP has the right approach, how about you give him 100 picks before you formulate an opinion...
Actually, even small samples can be used to form an opinion about the chances of a given capper having a specific record. In this case, the various bets are mutually independent (or at least very mildly correlated), we ignore pushes (not counted), and so the binomial expansion can be used given the handicapper's probability of success for a single event.
Assume Abundy's long term success rate is 60%. The cumulative probability, using the binomial expansion, of Abundy having 12 or fewer successes out of 28 trials is roughly 5%. In other words, we can say that the chances he is not a 60% capper are 19 out of 20.