Mark Buehrle, Strikeout Master?

Up until Mark Buehrle's June 24 start, he was having a typical Mark Buehrle season. He had thrown 92 innings in 14 starts (6.5 innings per start) and posted a 3.82 ERA and a 4.29 FIP that seemed pretty prototypical of most of his career. Perhaps most important to note was the fact that his strikeout rate stood at 11.8 percent, almost identical to his previous three-year mark of 11.9 percent. Everything seemed to be going about as well as one could have expected when the Marlins signed Buehrle to a four-year deal worth almost $58 million.

And then, out of the blue, Mark Buehrle started ratcheting up the strikeouts.

Starting in his June 24 start, Buehrle has accumulated 29 strikeouts in 28 2/3 innings. He has recorded at least seven strikeouts in each of his last seven starts. His strikeout rate jumped up to 26.6 percent. As you might expect, all of those strikeouts have helped tremendously; he has pitched at least seven innings in each outing and posted a 0.92 ERA and 2.02 FIP.

How in the world did Mark Buehrle do something like that? Well, Ben Duronio of FanGraphs has seen an interesting trend in changeup usage for Buehrle throughout the season.

The increased usage of the change up, specifically becomes even more magnified with two strikes.

Changeup:
0-2: 54.4% frequency, 5.6% whiff
1-2: 33.7% frequency, 10.9% whiff
2-2: 46.2% frequency, 11.7% whiff
3-2: 55.6% frequency, 20.0% whiff

Cutter:
0-2: 10.7% frequency, 9.1% whiff
1-2: 20.2% frequency, 12.1% whiff
2-2: 16.9% frequency, 4.5% whiff
3-2: 16.0% frequency, 23.1% whiff

With the new approach, Buehrle is netting a 4.06 strikeout-to-walk ratio, the first time he has been above three since 2005. His walk rate, which has been one of the biggest factors for his success throughout his career, is now at a career low 3.7% with 1.34 walks per nine innings pitched.

It is true that Mark Buehrle has been using the changeup more often this season, and that could be leading to more strikeouts. Duronio mentions that Buehrle's strikeouts have gone up since taking on this new seasonal approach, and that could very well be true.

But why has this only been seen during the last few starts? Prior to June 24, Buehrle's strikeout rate was identical to his rate from 2009 to 2011, indicating that he did not necessarily make many changes. Yet we have known about the increased changeup usage for a while:

Here are his pitch frequencies in 2012 from Brooks Baseball.

Buehrle, Pitch	%
Four-seam Fastball	29
Changeup	32
Cutter	20
Sinker/Two-Seam	11
Curveball	8
Slider	1

Now look at the frequency at which he used these pitches throughout the Pitch F/X era since 2007.

Buehrle, Pitch	%
Four-seam Fastball	28
Changeup	20
Cutter	24
Sinker/Two-Seam	16
Curveball	9
Slider	3

Most of his approach did not change, but we did know that the changeups have slowly increased over the last few years and that the changeup use had reached its high this season. But before June 24, there were no results to be noted. Perhaps this mystery deserves some looks before and after those strikeout starts began.

Taking a look at the raw Pitch F/X data (we usually use Brooks Baseball's re-adjusted pitch classifications when looking at seasonal data), we see little difference between the usage in the first 14 starts and the last four.

The usages really are not all that different. A few more fastballs, fewer cutters, and perhaps most importantly, two percent fewer changeups. If our theory that the changeup increase is behind the strikeout binge, than this does not seem to correlate well with that theory.

But Duronio mentioned that the increase in changeups was pronounced with two strikes, Maybe we could see a change in changeup usage between the two starts at the two-strike level.

It turns out that that is not the case either. Usage rate alone is not the issue, even with two strikes, as Buehrle used his changeup less in almost every count when he went on his recent strikeout binge. He also used the changeup on two-strike counts fewer times overall.

Contact

So I do not necessarily think it is usage. But there is something to be said about contact rates involved with the changeup. Before the June 24 start, hitters were whiffing on just about 12 percent of their swings at changeups from Buehrle. Since that start, it should not surprise you to hear that that rate has gone up to 15.0 percent.

How is that affected by a two-strike count, you ask? Buehrle whiffed batters at a 9.1 percent rate with his changeup on two-strike counts before June 24, but since that time, he has gotten them to swing and miss at a 17.3 percent rate. So we finally see something that favors the strike hypothesis, as Buehrle outperformed his overall swing-and-miss rate on changeups lately, leading to more strikeouts.

It is important also to note that an additional pitch saw a massive increase in swings and misses, and that is the curveball. Buehrle has made batters whiff on 16.1 percent of their swings on curves since June 24 as compared to 6.8 percent of those swings before then. His curveball is his second most useful "out" pitch, so this increase probably has a lot to do with Buehrle's high strikeout rate.

Conclusion

I am not certain exactly what has caused Buehrle's changeups to be more elusive in 2012 than in previous years. However, based on the fact that before the strikeout explosion starting in June 24 Buehrle did not significantly change in his strikeout rate of previous years, we can presume to a degree that his choice of changeups with two-strike counts likely does not have much to do with it. He has actually chosen to use the changeup in two-strike counts more throughout the season and especially before June 24, and yet the strikeouts did not arrive until June 24.

This has less to do with usage than it does with contact. Buehrle's pitches are evading hitters better, or at least they have been since June 24. It could simply be an awesome statistical blip in the radar, or it could have something to do with a tighter curveball in addition to the added changeups. But somewhere along the line, Buehrle went from being his typical self to a strikeout master, and a season-long change is not likely to have caused such a sudden, drastic shift midseason.