As I'm sure you know, expected value is all about estimating which bets are profitable and which bets are not.
Expected value = wager + (expected win - expected loss)
My personal picks (in particular in this thread) are games in which I can pick less than 50% and still come out with a profit.
My picks last night are the perfect example of this.
I'll even use a game tonight as an illustration of what I mean:
Blackhawks v Predators. In my opinion, if the Blackhawks play the Predators in Chicago 100 times, they will win by 2 goals at least 50 of those times....at 2.05 odds, to me, the long-run average of the results will be in my favor and thus eliminate the house edge of the bet.
In sports betting, there aren't exact calculations as to how many times, in the future, team X will beat team Y by 2 goals or by 2 runs...but if you find yourself betting plays you think are 50% likely to happen and you are getting odds at over 2.00 (+100), then you should make money.
Again, this is all opinion based...the general premise of the thread is that, I can quadruple your bank roll by flat betting games that are usually over 2.00 odds but that I think are likely to happen at least 50% of the time...