That's impressive, but you are going to need a much larger sample. I did two recent studies that were profitable with spot plays covering much larger samples (not with longshots) that fell apart the next year. Overall the result are still good, but when things fall apart you can't tell if the early part of the sample was a fluke or the current part of the sample is a fluke.
__________________
"Unlearning is the highest form of learning"
|