Baseball Prospectus / MLB.com Playoff probabilities

[HelenaEngineer]

I don't mean this as a "we aren't getting any respect" vent but am really trying to understand the modeling that could yield the results Baseball Prospectus (on MLB.com) shows for playoff probabilities. For example - the current postseason probabilities for AL East teams are shown as Tampa Bay 64, New York 55, Toronto 48, Baltimore 24 and Boston 11. The Orioles are effectively tied with Toronto and 1/2 game behind New York and 7 games ahead of Boston - yet the difference between the Red Sox and Orioles is much smaller than the gap between Baltimore and Toronto or New York.

The Orioles have a better run differential than New York (+41 to +26) while Boston is -45, yet the Yankees are more than double the chances of making the postseason. The Orioles have recently added Wieters, should have Hardy for the rest of the season after he missed a chunk of games and had a slow start due to the layoff, and will bring back Schoop soon - so their 2B/SS/C production in the last 90 games should be far better. The Yankees are ancient and have really stressed out their bullpen with poor starting pitching.

I get it that there might be different opinions, but can an objective observer really think the Yankees have a far better chance of making the postseason than the Orioles - and if so based on what?

I can see one area of concern that the Orioles have had a very batting average with runners in scoring position and have allowed a very low average, and those two stats are not likely maintainable.

Am I just a homer, or am I correctly identifying a deficient model that doesn't make a lot of sense? Just because they run a ton of Monte Carlo simulations does not mean their input assumptions make any sense. As a statistician, if I saw this output from a model I had built, I would be concerned about its validity and look for the flaws in my assumptions.

Helena Engineer