A possible red flag for Bumgarner in free agency

Madison Bumgarner has spent his entire career pitching for the Giants, and you already know what that resume looks like. There are the three rings, the four All-Star appearances, the legendary postseason moments, and while you're surely thinking about that 2014 Game 7 relief outing, don't forget those eight shutout innings in Game 4 of the 2010 World Series just weeks after he turned 21 years old. There were also the pair of fluke injuries in 2017 and '18 before a return to form in 2019, but you get the idea. He's a solid-to-somewhat-above-average pitcher with the best October pedigree you can imagine.

That's all likely to get the 30-year-old lefty a three- or four-year deal at something like $15 million to $18 million per, which is all well and good, but it does open up a pretty large question. Go back to that very first sentence, where we pointed out that San Francisco is the only baseball home he's known. As everyone knows, Oracle Park has long been one of the most pitcher-friendly parks in the game, to the point that Brandon Belt probably has nightmares about "Triples Alley." The Giants are moving toward shifting the location of the bullpens from foul ground to the outfield this winter, and a side effect will be the distance of the fences.

What does all of this mean for Bumgarner, should he end up somewhere a lot more hitter-friendly? The reason we ask is because it's difficult to overlook just how much better he's been at home, both this past year and over his career.

Bumgarner, 2019

Home -- 2.93 ERA, .261 wOBA (2019 Major League-average wOBA: .324)
Road -- 5.29 ERA, .349 wOBA

Bumgarner, 2017-2019

Home -- 2.62 ERA, .264 wOBA
Road -- 4.84 ERA, .343 wOBA

Bumgarner, Career

Home -- 2.72 ERA, .270 wOBA
Road -- 3.53 ERA, .300 wOBA

This is not to imply that he's merely "a product of a favorable home field," because surely he's more accomplished than that. No one really believes that if Bumgarner were a Twin or a Cardinal or a Brewer, he'd post a 5.29 ERA over a full season. Still, it's our job to figure out just what's going on here. Is it really just balls that would been homers elsewhere turning into outs in the deep San Francisco outfield? Probably not.

The best way to do this is to separate what happens when the ball is hit from when it's not, so let's do that.

Did his home field help him with strikeouts and walks?

Answer: Maybe, but not a lot.

We should clarify here that it might not be the field, exactly, it might be that he likes his own bed or his regular coffee shop down the street, but we're going with what we've got. People generally consider "home field advantage" to be about altitude or fence distance, but as we've explored with Detroit in the past (and as Eno Sarris once wrote specifically about San Francisco), simple things like the angle of sunlight and the batter's eye can affect non-contact outcomes, too.

Over his career, Bumgarner has a 25.3% strikeout rate at home, and a 22.6% mark on the road. His walk rate is about the same at both. That's not nothing, but it's not by itself going to change his outcomes.

This, mostly, isn't it.

Did he simply allow louder contact on the road?

Answer: Not louder, but maybe higher.

Maybe a home-field advantage goes beyond just making contact or not. Maybe it goes toward making quality contact. We've got Statcast data extending back to only 2015, so we'll stick with the last five seasons on this one.

Since the start of 2015, he's allowed a hard-hit rate of 34.5% at home, and ... 34.5% on the road. (That's an identical exit velocity of 87.8 mph both home and road, if you prefer.)

So there's that, but there's more to this than just how hard the ball is hit. He's allowed 38% grounders on the road over the last five years, and 42% at home. The balls allowed in the air haven't been hit any harder -- identically, home and road, on flies and liners -- there have just been more of them. Balls in the air (obviously) cause more damage than grounders.

If we wrap in launch angle to the exit velo, we do see a little bit of an effect. To get to that, we'll use a metric called "Expected wOBA" to get to quality of contact. (We're using a version here that is looking only at batted balls, not the usual all pitches.) It's specifically not adjusted for park, which is the point: It asks how damaging the contact allowed is.

At home, since 2015, he's allowed a .352 ... and on the road, it's .385. That's actually kind of a big deal. Just in 2019, the difference was .371 at home and .427 on the road. That latter number, the .427 expected mark on batted balls on the road, was the fourth-highest of 125 pitchers who allowed 150 batted balls on the road.

This is both telling -- he truly got crushed on the road this past year -- and somewhat unsatisfying, because if the idea here is that perhaps the big dimensions of San Francisco helped him, well, we haven't even made it to that yet. Just based on quality of contact, road Bumgarner in 2019 was a below-average pitcher. That doesn't guarantee he will be, but that he was.

Did the contact he allowed just hurt him less at home?

Answer: Definitely.

OK, now we're talking. This would be where the dimensions of Oracle Park would come in, like if, for example, a certain left-handed pitcher allowed a ball that would have been a home run in 26 of the other 29 Major League parks -- as the announcers say, "that's in the pool in Arizona" -- but it settles harmlessly into Kevin Pillar's glove for an out.

(That just looks like a regular fly, doesn't it? That's the beauty of San Francisco. But if you superimpose the trajectory over, say, St. Louis's Busch Stadium -- the Cardinals are rumored to be interested in him -- you can see what we mean.)

An easy way to do that would be to simply compare his expected outcomes on batted balls against what actually happened, especially since we already have that info handy. That is, we can start from the baseline of "somewhat harder contact allowed on the road" and then go from there, not just looking at the overall outcomes.

On the road, Bumgarner has more or less received what he's earned, with his actual outcomes on batted balls bouncing around ever so slightly on either side of his expected, with the total being a mere 5 points of wOBA. (That's .385 expected, and .390 actual. It is essentially nothing. This means that Bumgarner, on the road, over the last half-decade, hasn't really been the recipient of good or bad fortune or good or bad defense. He's just been what he's been, for better or worse.)

Now let's add in what happened at home to this chart. Every single year, by a lot, his actual outcomes have been better than his expected outcomes. The total here is a full 39 points, the distance between his .352 expected and his .313 actual. Among the 188 pitchers with 500 batted balls at home in the last five seasons, the gap is fifth-highest. If that's not entirely park effects, it's close, because three of the four names ahead of him have primarily called big West Coast pitchers' parks in Seattle and Oakland home, those being Kendall Graveman, Daniel Mengden and Felix Hernandez.

Again, in the same way you can't just look at "road stats for Rockies hitters" to assume you know what will happen to hitters leaving Coors Field -- DJ LeMahieu ought to have proven that in 2019 -- you can't just assume that a non-San Francisco version of Bumgarner will be poor. You can just assume that it will hurt him, by some amount.

Bumgarner, right now, may or may not be the best remaining second-tier free-agent starting pitcher remaining, among a crowded group behind Gerrit Cole and Stephen Strasburg. You might, for various reasons, prefer Hyun-Jin Ryu or Dallas Keuchel, now that Zack Wheeler is headed to Philadelphia. Any of those opinions are reasonable, and giving Bumgarner a mental boost for what he's done in October is reasonable, too.

It just might not be exactly the same Bumgarner we've seen. San Francisco is an absolutely wonderful place to pitch, possibly the best in baseball. He won't have that on his side anymore. It's an open question as he (likely) heads to a new home.