The first Vegas sportsbook released their over/under win totals on each NBA team this season. The totals can be seen here on the ESPN page it was mentioned on:

http://espn.go.com/espn/chalk/story/...eague-best-585

Observe that the sum of each team's wins comes out to exactly 1242. Considering the fact that there are 1230 wins to be distributed among the 30 NBA teams leaves us with the conclusion that a majority of these teams *should* finish with a win result that is "under" their total (assuming of course that the expectation of variation of their win totals are distributed equally).

Now, ignoring juice, how would one calculate how much of an edge they're getting if they bet the under on each team? Considering the fact that only 1230 wins can happen within an NBA season, along with the fact that the sum of each team's win total at this book equals 1242, means that in the long run, betting the under on every team should result in a profit, right? (ignoring juice obviously).

There has to be an edge here, but what is it? And how do you quantify it? Distribution and variance experts are especially welcomed to answer this.