Posted this at CM so might as well post it here:

Favorites are favored because they have been designated as having a higher probability of success. Therefore, assuming all the market vig is proportional to the outcome, finding a better number on a favorite than what the market has at fair value odds (or just getting a better number than pinny), should this not equate to long term profit?

Let’s say Pinnacle has a line on an MLB game set at -130 for the favorite, and +122 for the dog

Adjusting the implied probability to fair value probability (discussed here), the probability is roughly 55.6% of success.

The expected value of this wager is

EV(X) = .556(1) + .444(-1.3)
EV(X) = -2.01% expected loss

Some other sportsbook that has gotten lazy with market awareness is offering the same team at -120. Therefore we have a team with probability of success marked at 55.6% with -120 odds.

EV(X) = .556(1) + .444(-1.2)
EV(X) = +2.4% expected return

Stating the obvious, yes, but when laid out in presentable form the concept is much more enlightening.