Adrian Gonzalez’s Evolution, Part II

Part two of a two-part series.

In Part I of this series, I dove into some Pitch F/X data to try and tease out how Adrian Gonzalez changed his approach after leaving San Diego in 2011. Overall, pitchers did not appear to adjust their approach to Gonzalez, evidenced by the fact that the distribution of pitches by location in 2011 was almost identical  to 2010. Gonzalez, however, did seem to alter his approach by altering what pitches he offered at, most notably swinging at more balls away and up in the zone. The change in his swing distribution combined with the change in his performance seemed consistent with the theory that the slugger was purposefully being more aggressive on pitches outside of the strike zone in an effort to take advantage of the Green Monster at Fenway Park. But to get a firmer handle on this we needed to split out Gonzalez’s data by home versus away.

That is the focus of this article.

Obviously, the change in Gonzalez’s overall performance can be explained in large part by Park Factors. Yes, his wOBA was ~20 points higher last year, but he moved from a park that was brutal to left-handed hitters to one that was far more favorable. That being the case, we might have expected pitchers to approach Gonzales differently based on what park he was playing in.

Here’s a comparison of pitch location, year-over-year, by home and away:

We see some variation between how pitcher’s approached Gonzalez at Fenway versus Petco, but the difference isn’t all that drastic. The areas that show the most contrast are pitches up and away, as well as up over the middle of the plate. Pitchers threw less pitches up and away, but challenged him more high and over the plate. Gonzalez was roughly even in those two zones at home last year (.059 wOBA/swing high and away, .042 high over the middle), but I would imagine pitchers in 2011 were hedging against Gonzalez pounding balls off the Green Monster.

In terms of away ballparks, pitchers seemed to challenge Gonzalez away much more than on the inside part of the zone. The percentages are not huge, but a pattern does appear where pitchers decided to avoid coming inside to Gonzalez more so than in 2010.

Did Gonzalez adjust as well? Generally, yes.

Gonzalez significantly increased the percent of pitches he swung at at home overall. However, we see that he swung at 11% more pitches up and away, 7% more down and away, and 13% on high pitches over the plate. Now, there isn’t a one-to-one correlation between pitch location and hit location, but Gonzalez’s approach is consistent with trying to go the other way more at home.

The changes were observable, but less drastic on the road. Gonzalez increased the percent of pitches he swung at on the outside part of the plate, taking advantage of what pitchers were giving him. He generally avoided more pitches up and belt-high on the inside. However, Gonzalez drastically increased the balls he offered at down and in (+17.5%) while on the road. He didn’t reap a huge reward for doing so (wOBA in that location only increased .25 on the road), but he did see an increase nonetheless.

Gonzalez did seem to intentionally take greater advantage of a friendly left field at Fenway, and this really pops when we compare the difference between his swings at home versus away in each year:

The two graphics above show Gonzalez’s swing percentage in each zone at home minus his swing percentage in the same zone away in 2010 and 2011, respectively. What we see is that the difference on the outside of the plate was more drastic overall in 2011 (+10% Home/Away split in 2011 versus 2010), suggesting again that Gonzalez was very intentional about trying to leverage the dimensions of Fenway.

There is still the question of whether Gonzalez’s changing approach accounts for the change in his performance, or if it was simply a function of playing his home games in a more hitter-friendly park.

Let’s move away from the heat maps for a moment and just look at the change in Gonzalez’s wOBA and BABIP at home and away.

Based on the data I had, I calculated Gonzalez’s overall wOBA by batted ball type and then the percent change from 2010 to 2011:

wOBA	2011	2010	Diff	% Change
Flyballs	0.637	0.497	0.139	28%
Groundballs	0.229	0.184	0.045	25%
Line Drives	0.794	0.734	0.060	8%

Now, overall his wOBA increased for every batted ball type, with flyballs and ground balls seeing the biggest jump. This makes sense, since line drives aren’t likely to increase all that much. The real question is whether the increase was more drastic at home versus on the road.

wOBA – Home	2011	2010	Diff	% Change
Flyballs	0.597	0.434	0.163	38%
Groundballs	0.218	0.178	0.040	23%
Line Drives	0.818	0.763	0.055	7%

wOBA – Away	2011	2010	Diff	% Change
Flyballs	0.667	0.555	0.112	20%
Groundballs	0.240	0.190	0.050	26%
Line Drives	0.768	0.717	0.051	7%

What should jump out at you is the difference in wOBA per fly ball. Gonzalez picked up an extra 38% of wOBA in Fenway last year, 18% more than his improvement on the road. However, he did increased his wOBA on fly balls 20% on the road between 2010 and 2011, which is hard to simply chalk up to park factors. So while Gonzalez did likely get a boost from playing in Fenway, there is suggestive evidence that it wasn’t the lone variable accounting for the increase in production.

To wrap things up simply, Adrian Gonzalez did appear to alter his approach to take advantage of Fenway Park in his first year in Boston. Gonzalez was more aggressive at home than away when it came to outside pitches, suggesting that he made a point of attacking the Green Monster with line drives and fly balls.

From a performance standpoint, this approach appears to have paid off. Gonzalez increased his wOBA per fly ball by 38% at home in 2011. He also increased his BABIP at home by almost 100 points. Now, the degree to which that is sustainable is certainly unlikely, but if Gonzalez continues to take advantage of Fenway’s dimensions we would expect his BABIP at home, particularly when going the other way, to be consistently high.

We should know more about the sustainability of this change this year. The current analysis is working off of only one year’s worth of data in Boston, and we don’t have a good sense as to what degree of variation in swing distribution is likely to be random year to year. But the findings are directional and consistent with the notion that Gonzalez did tailor his approach to his new environment, and that approach contributed in some way towards his increased production.