[FreeFall]

This is a creative idea I had the other day and was curious what you guys thought. Say I tracked a bunch of diff random variables. Like all NBA tots, Soccer tots, and MLB tots (notice these are all "50/50" lines). Then say we make bets only when they are 0-9. Is this a foolish idea? Is there any advantage to betting on something only because it is far from the mean?

[bztips]

Yes, a very foolish idea.
They're independent events, the point total in a given NBA game has nothing to do with what the point total in a prior NBA game was (there's no "memory", just like we always say to those who think they should bet a certain lottery number because it's "due").

[FreeFall]

Yes, a very foolish idea.
They're independent events, the point total in a given NBA game has nothing to do with what the point total in a prior NBA game was (there's no "memory", just like we always say to those who think they should bet a certain lottery number because it's "due").

I'm not disagreeing with you, but not agreeing either. Say for example you found a edge of 2% over 1000 games. you are 0-990. That means there is no reason to bet more on the last ten games of the set? I think I'm miss some underlying concepts.

[bztips]

Well, first of all I'd seriously reconsider if my .02 EV was really any good if it just lost 990 times in a row. But ignoring that, YES there is no reason to bet more on the last ten games just because you've gone through a losing streak. Now there may be OTHER reasons to change your bet size (a la Kelly), but they have nothing to do with the question that you asked.

[Hybris]

http://en.wikipedia.org/wiki/Gambler%27s_fallacy

[Wrecktangle]

Take an infinite amount of monkeys typing on an infinite amount of typewriters, sooner or later a monkey will write Hamlet. Somewhat like going 0-9, or 9-0.

Now, would you want to hire that monkey to write Romeo and Juliet for you?