Wow.  Those are solid lines.  The only club that I have more than 5 games outside their projection is Tampa over 67.5.  Even that line is not really unreasonable except in the sense that the odds of Tampa winning more games than that is probably twice the odds of them winning less.  Usually there are three or four lines that I feel relatively confident about.

I'd take the Yankees under and the D'Backs over, as well as the D'Rays over.

Last year I had the Yankees under, and it took the Jays beating them on the final day to keep them in the under column.  Go Lind!

Notice that a lot of teams have win projections in the 80's?  There is an interesting arithmetic error they have made in setting the lines.  They have projected too many wins.  The average number of wins per team will be 81.  Therefore, the total expected number of wins is 2,430.  These lines project considerably more wins than that.  If you pick the under on every team, you'd have an excellent chance of coming out ahead, depending on the dispersal of actual wins from team to team. 

There's no arithmetic error - the line is set based on where they feel they can get equal money bet on one side (or money bet on the wrong side).

And the average wins of the lines above is 81.7.

My point is that the setup is such that one could beat the system.  Also, if you are just looking at the numbers to see how many wins they think your team might get, keep in mind that they are overstating the total possible wins by an average of almost a win per team.  That is the arithmetic error I am referring to.

0.7 games above average is less than a 1% margin to exploit.

The house is getting 15% to take the bet.

15% plus 7-8 months.