Ohiopiper and I got the exact same "analysis". You really have to stretch the results to fit them to me. I give it 10%. Pretty poor showing for random, where I would expect closer to 50%.