Fences Coming In

The last game of the season just happened, and that feels like a big enough deal to write about, but is it really? Neither the Mariners nor the Angels had much to play for, although the M's seemed to care a bit more with their convincing 12-0 victory. They beat Jered Weaver in the process, dropping him to 20-5 on the season. Fun fact: 3 of Weaver's 5 losses in 2012 were to the Mariners. Go figure.

Anyway, the more meaningful news is that the Safeco Field's dimensions will be different in 2013. The fences are coming in, particularly in left-center field. The changes should make hitting easier, but what kind of an effect will it have?

Let's start with quantifying how cavernous Safeco Field played in 2012. The Mariners had a .275 wOBA at home, easily the worst mark in all of Major League Baseball. 7.7% of their fly balls at home went for home runs, easily below the MLB average, but still better than the Padres, Marlins, and Giants.

Of course, not all team lineups are created equal. The Mariners have had some rough seasons in Safeco, but been just about as atrocious on the road. That suggests something more about their lineup than their ballpark. 2012 was a different story though. The Mariners had a .305 wOBA on the road and 11.9% of their fly balls went for home runs. The differences are noteworthy, but wOBA is particularly shocking given that MLB teams as a whole posted a .323 wOBA at home this year, but just a .308 wOBA on the road.*

*Percentage of fly balls that were home runs had a negligible difference, 11.9% at home and 11.5% on the road.

So what exactly will the new walls do? Below is a diagram of Safeco Field with balls in play data from 2012 (both Mariners and opponents). I've only included doubles, triples, and fly balls. The yellow lines are the new walls sketched to the best of my ability. Ball in play data doesn't have the incredible precision that the scatter of dots might suggest either. In other words, the picture gives a feel for what might have happened this season with the new dimensions, but no definitive answers. See what you think:

image from http://katron.org/projects/baseball/hit-location/

As you can tell from the 420-foot "double" well past the fence, ball in play data isn't perfect. Still, only the phantom double and another double in the left-center field gap clear the new dimensions. There are a few other balls, maybe upwards of 20 if we want to be really liberal, that might clear the new fences. Remember, this number represents home runs for every Mariner and Mariners opponent combined over a full season.

We should see a few more home runs in Safeco Field next year, but I am not convinced that the park will play much differently. Even if there are 20 more home runs total generated by the new fences, that would be 1 more home run, for both teams combined, over the course of 4 games (a long series). That would make a difference, but I doubt it would turn Safeco Field into a different ballpark.

I would not be surprised to see offense improve next year though, thanks to a number that isn't getting much attention. The Mariners BABIP at home in 2012 was a paltry .262, easily worst in baseball. The year before they had a .279 BABIP at home, and in 2010 it was .289. On the road this season it was .287. There is a case to be made that the M's were unlucky at home this season. A fair amount of balls they hit did not find holes that usually do. 

The BABIP issue is one I looked into much further. I took home/road splits for every team in baseball from 2000-2012, since 2000 was Safeco Field's first full year, and examined the differences. Bars pointing upward represent teams with higher BABIP at home, and bars pointing downward represent worse BABIP at home. The M's are the teal bar, the rest of baseball blue:

The two extremely tall bars on the far left of the graph belong to the Rockies and Red Sox, respectively.

Colorado's home, Coors Field, is known for giving up more home runs than most parks, but the BABIP difference is even more pronounced. This makes intuitive sense because the ballpark's outfield dimensions are the largest in baseball. Even though balls fly farther in the mile-high air, outfielders do not cover much more ground. More balls are going to fall in with more outfield space, hence the elevated BABIP.

There really isn't much that can be done to make Coors Field more neutral. The fences could be moved back further, and that would lower the home run rate, but it would also cause the BABIP to rise even more. The Rockies could make offense look different, but not necessarily better or worse.

The Red Sox play in legendary Fenway Park, where the Green Monster dominates left field. Interestingly, the Red Sox have actually hit home runs more frequently on the road over the last decade, but again, this might make some intuitive sense. Right field and center field are actually rather deep in Fenway Park, and a number of well struck balls out to left field don't get over the Monster for dingers. The BABIP data suggests to me that the Monster also takes many routine fly balls and transforms them into singles, or, maybe the space in center and right field account for the elevated BABIP by itself.

Buttressing the Mariners on the graph are the Giants, Dodgers, and Athletics to the left, and the White Sox and Padres to the right. The White Sox and Dodgers pair a lower BABIP with an elevated home run rate, suggesting that their parks have smaller outfields. That could explain why more home runs leave the yard, yet more balls in play turn into outs. However, the Giants, Athletics, and Padres (along with the Mariners) all have lower home run rates at home than on the road. This is odd and counterintuitive - somehow, these outfields play like they are small according to BABIP, but big according to home run rates.

The Giants, Athletics, Padres, and Mariners also all happen to play in ballparks near water at sea level near the Pacific Ocean. This could be a coincidence, but I doubt it. All of these cities seem to regularly have rather cool evenings, thanks to an ocean that naturally cools the cities off. Baseball go farther in warmer air, particularly warm, muggy air. The west coast isn't noted for its humidity like the south, and to a lesser degree, the east. The nature of the air in these ballparks might be the source of their pitcher-friendly environments.

I suspect that the Mariners can't do much to change the nature of Safeco Field, much like the Rockies can't do much more to Coors Field. The M's can move the fences in and generate more home runs, but that could come at the cost of even fewer balls dropping in. Much like Coors Field will always be a mile above sea level, Safeco Field will always be subject to an evening onshore flow.

Home runs get all the attention, and those are suppressed at Safeco Field. They should be suppressed less in 2013 with the fences coming in. However, BABIP gets overlooked, and it is at least as big of a factor for the lack of offense. Less hits mean less baserunners mean less runs. Moreover, the presence of fewer home runs and fewer hits suggest to me a larger, overarching factor that might not be addressed with some simple fence manipulation. Offense could look different at Safeco Field next season but the fences do not guarantee improvement.