By Jason Wojciechowski on August 6, 2013 at 8:35 PM
I'm not usually an instrumental rooter, someone who picks the non-A's teams to cheer for on the basis of who helps the A's most, except in narrow, obvious circumstances like "if the Rangers lose today, the A's go to the playoffs, hooray hizzah." I don't generally root against the Rangers, for instance, even though over the course of the season, given how the Angels have turned out, the Rangers are going to be the team who keep the A's from the playoffs if anyone does. But I like the Rangers, I like how they're built, I like that they hired Ron Washington. Making the playoffs is easy. Tell 'em, Wash.
But sometimes I like to think about who would help the A's and who wouldn't, who I'd prefer the A's face in a playoff series and who I wouldn't. We do this a lot more in basketball, analyzing strength vs. strength of the two teams and figuring out which matchups might be best for your squad. In baseball, though, the versuses are too complex and there's too little data, so we mainly just hope to get the team missing their ace or the crew with the hobbled first baseman.
One aspect we can think about is parks. The A's have an extreme fly-ball pitching staff. From 2004 to 2013 (i.e. the last decade), per FanGraphs, zero teams' pitching staffs have allowed fewer ground balls (as a percentage of balls in play) than the A's staff. This works for the A's in part because of their outfield defense:
|Rank||BABIP allowed, non-grounders|
|2||Tampa Bay, .326|
"But that foul ground!" The data I'm working with, provided by TruMedia (and probably relying at bottom on MLB Advanced Media, though that's not clear), doesn't allow me to actually remove popups, unfortunately, though I will note that the A's rank just ninth in baseball in total number of "infield air outs" according to TruMedia, and we're talking about around 5 percent of all balls in play, so it's not having an enormous effect on the overall rank.
It also works because of their park. Using StatCorner's rolling three-year component park factors (which, as you know, are scaled to 100 such that an 80 means that the park reduces the component by 20 percent and a 115 means it increases it by 15 percent), Oakland's home run factors are 71 for left-handed hitters and 93 for righties. (I know I moved from balls in play to homers, but I'm not doing a study here. We're just yakkin'.)
Here are the American League teams in serious playoff contention, their home run park factors, and their current Baseball Prospectus adjusted playoff odds (i.e. their odds of getting to the Real Playoffs, not just wandering into that one-game wild card nonsense) before today's games, rounded to the nearest whole percent.
|Team||Playoff Odds||LHB HR||RHB HR|
So the ideal situation is the A's having the best overall record, Boston winning the East, Detroit the Central, and Tampa Bay knocking out any of the other three teams in the Wild Card. This way the A's get the Rays and their suffocating home park in the first round while never having to deal with Texas or Baltimore's nonsense.
Cleveland would not be so bad for Tom Milone, but with AJ Griffin, Dan Straily, Bartolo Colon, and Jarrod Parker all possibilities to start in the playoffs (or, really, with Colon and Parker locks and Griffin and Straily possibilities), I'd take the park that stops lefties (Boston) over the one that stops righties (Cleveland).
Assuming Bartolo Colon is the team's ace heading into the hypothetical playoffs and assuming that Brett Anderson is not in the rotation, it will be hypothetically theoretically interesting to see how Bob Melvin orders his rotation. We all love to shout and moan over Tom Milone's record at home and on the road, but if the A's were to play Tampa or Cleveland, their home run factors for righties are actually lower than Oakland's, so it might make some sense to let Milone take a shot at a road start (to the extent the A's have the luxury of lining up their rotation, of course).
(I can't even begin to list all the other caveats I should have mentioned: I'm assuming that StatCorner's methodology is sound, that the factors reflect reality, that October in these parks is comparable enough to May through August to make decisions based on regular-season data, that park factors in general are even methodologically and philosophically sound. I'm missing some, I'm sure.)