What We Can Say About Safeco Field
There were times this summer — lots of them, in fact — when I couldn’t tell if I cared about baseball anymore. The Mariners weren’t provoking any sort of emotional response at all, and that’s supposed to be my most favorite team. I didn’t just not want to watch them; I actively avoided them, and I recognized my own behavior. But it turns out it wasn’t baseball — it was the Mariners. The Mariners were downright unwatchable for stretches. I’m sitting here, watching the playoffs, and it’s incredible. No one has a better atmosphere than the Pirates right now. Pirates fans have been through a lot worse than we have. The Pirates are proof that the Mariners can be good again, and Safeco can be packed and loud again, and baseball can be fun again. Don’t know when, but no matter how you feel about the Mariners today, they’ll pull you back if and when they can win 90 games. Baseball’s awesome when it matters.
The Mariners have a vested interest in mattering, and last offseason, a step they took toward that end was bringing in most of the fences at Safeco Field. It wasn’t something that was going to directly give the Mariners any more wins — it was a change that would affect the Mariners and their opponents equally. But the change was intended to make the park more neutral. More fair. More appealing to other players. More tolerable for current players. There probably was too much imbalance before, and it’s possible the park got into some hitters’ heads. Safeco seemed overdue for a fence-bringin’, and over the course of some months those fences were brought. A full season, now, has been played.
So it’s only natural to wonder what’s become of Safeco Field. We grew accustomed to all the old park effects. How about the new park effects? We should have an idea by now, right, since it’s been a whole season? What is Safeco Field, v2.0? The answer is not contained within this post. But there will be numbers anyway.
The most important thing to get is that park factors take more than a year to overpower the noise. There’s signal in there — there’s signal in there after even just one single game — but this is going to take years. Split the season in half, first. That’s how many home games there were. Then you have to consider all the balls in play, and how many of them were grounders, or routine flies, or homers that would’ve easily cleared any fence. Only a small percentage of balls in play are subject to park effects, and there aren’t that many of them hit to each location. Plus, there’s the matter of other, unseen, confusing park effects, like effects on walks or groundball rate. Parks do lots of stuff, directly or indirectly, and you just have to let the games be played for a while.
There are a few things we know for sure, though. The fences were brought in in left, left-center, center, and right-center. The fence was lowered a little in left. Automatically, we know that’s going to mean more dingers. Reasonably, it can’t not. The dinger threshold is reduced, so home runs will go up, because they have to. The question is by how much, and the other question is what else is going to happen? What happens to doubles and triples? What happens to run-scoring, overall?
Nothing can be said conclusively, but we can at least look in the numbers for clues. Below I’m going to include a table, showing 2000-2012 Safeco Field, and 2013 Safeco Field. The percentages are simple park factors. A percentage of, say, 90% means that Safeco’s rate was 90% the same measure on the road. For example, between 2000-2012, games in Safeco had a .252 batting average, while Mariners road games had a .269 batting average. The percentage shown is (.252/.269) * 100 = 94%. This approach is too simplistic, but I have interest in simple approaches.
For the record, the 2B/3B and HR rates are rates on contact, not rates overall. Anyhow, to just walk through: this year, batting average was almost even. OBP was almost even. BABIP was almost even. Doubles and triples were up a lot, combined, while homers didn’t actually budge. Something might’ve happened with walks, but that’s going to take years to sort out, just like everything else. There’s so much noise. All of these error bars overlap the previous park factors, so none of this, I have to imagine, is statistically significant.
But, there you have it. Safeco didn’t quite play neutral in 2013 — OPS in the park was .705, and .727 elsewhere — but it played more neutral, especially compared to the season before. Interestingly, this didn’t happen as would’ve been projected. Instead of increasing home runs, it left home runs the same and increased other hits, despite reduced in-play surface area. This is the opposite of what we would’ve thought, and compelling evidence of just why we’re going to need a multi-year sample. These results don’t quite make sense, and there’s no reason to believe Safeco isn’t going to increase homers going forward from last winter.
One other thing I did look at was ESPN Home Run Tracker data, going back to 2006. One of the things that’s measured, for each home run, is how many ballparks would’ve allowed that home run to be a home run under standard conditions (no wind, room temperature). The maximum value, of course, is 30. The minimum value is 0, in case the homer got help from the air in some way. Between 2006-2012, Safeco averaged 19 homers a year that would’ve left 15 or fewer ballparks. It averaged eight homers a year that would’ve left ten or fewer ballparks. Last season, there were 37 homers that would’ve left 15 or fewer ballparks, and 18 homers that would’ve left ten or fewer ballparks. So there’s some evidence of more hitter-friendly dinger conditions, and it goes along with our own memories of some homers, especially to left-center, that wouldn’t have been homers before. Ryan Divish has casually estimated that Safeco yielded ten extra homers this season, and that seems to be roughly in the ballpark. That’s not meant to be a play on words, but I don’t care enough to think of another expression.
So Safeco played differently in 2013, as expected, but it didn’t play differently as expected. More than anything else, that’s a demonstration of the noise that goes into single-season park factors. As for how Safeco feels, now? It probably still feels big, especially on cold nights, but hitters should no longer live in such fear. Don’t know if the new Safeco is going to mean brighter futures for the Mariners’ young position players, thanks to increased confidence, or decreased non-confidence. At worst, it ought to have no effect, so there’s no reason to be upset. With the park. Lots of reasons to be upset with the Mariners.
Something I neglected to mention earlier: anecdotally, it seemed like a long and dry and beautiful summer. Certainly, it was a better summer than the two previous summers in Seattle, and that would’ve had an effect both on the gameplay and on the attendance. There’s also the matter of what might have changed as a consequence of the new big giant videoboard, but that’s just going to be folded in with the rest of the ballpark adjustments. We don’t care so much about the individual components; we care about the overall park factors, of the park as a package, and as more and more time passes, we’ll get a better and better idea of what Safeco has turned into.
For now, we’re left with assumptions and approximations. We know that Safeco has changed its clothes, but we don’t quite know what it’s decided to wear. It’ll emerge from the dressing room slowly, over a matter of a handful of years. This has been one godawful metaphor.