As you can see I havent contributed to this site very much. I was basically trying me own research and came up with something I would like to share here. How I make my picks is a mix of sabermetrics, some chase characters as well as situational plays. I backtested the chase part back to the 2004 season and it payed off very nicely in each and every season. Over the course of the last 5 seasons that the worst season still made 54,6 Units and I played it the first time in the 2009 season year and made 108,73 Units.
There is one thing I want to point out first, since a lot of people feel uncomfortable with betting on underdogs. The big difference in baseball compared to other professional sports is that the MLB is very close and the best and worst teams are not that different (witch respect to the win%) when compared to other sports. Taking into account that the baseball season has the "special item" that the games are played in series (3 or 4 games at one place in a row) and that only in 10-15% of these series the home team looses all three or four games, made me suspicious (please excuse my english, I am from Germany and doing my best ;-)). Taking out stinkers like Pittsburgh and Washington during the last years it looks even better. What I do on the chase part is the following: I divide the league in 3 groups with the best teams being in group 1, the middle in group 2 and the worst in group 3. The basis for the dividing is the win %age of the teams. Based on that you can decide which teams qualify for a chase and which dont. I dont want to go too deep with my approach as of now but I think I gave you an idea how to go from here.
With respect to sabermetrics I basically use available statistics from baseball-reference.com and put them into my excel-model to find out, what the outcome of this game should be. If I find a line that looks weak with respect to my outcome, I take the chase part into account as well. If this still fits, I take a closer look at the situation and some other influences as time, weather, rest etc.
In total I think I can say that my approach is definitely more statistical driven and the math behind the game drives my picks. I am looking forward to see how it pays off this season and am happy to discuss my approach. Feel free to ask questions or give comments!