A pitching quest for the perfect balance

Like yours truly, baseball contains a plethora of attractive features. There's the symmetry of the diamond, time conceptualized by a number of outs, an isolated battle between a pitcher and a hitter, and of course, that sweet crack of the lumber when good contact is made. This list could clearly go on much, much longer. But baseball has another feature that I find particularly inspiring: change.

The game is not static. Instead, it exists in a perpetual state of fluctuation, as teams and players are always attempting to maximize their opportunities for success. I love this aspect of baseball, because if you're a cerebral fan (and let's face it, you're at Beyond the Box Score so you probably fall into this category), there's always something to be tracking. Baseball is beautiful insofar as you get out of it what you put into it, in a way. If you want to drink a beer and just take it all in, be my guest and enjoy your time. If you want to study the details of what's constantly evolving under the surface, well, keep reading.

Every baseball play starts with the pitcher, and pitchers are always looking to improve their odds of defeating the batter in the box. Their weapons to do so are contained in an arsenal of pitches that they have at their disposal. But here's the thing: For most pitchers, their pitches are not equally effective. They may have one pitch that is vastly superior to their other options. And to make matters more interesting, the effectiveness of pitches can wax and wane over time as the feel for a changeup grows or, perhaps, the velocity on a fastball declines. In essence, it's always changing.

The 2014 season was no exception to these trends as many pitchers altered the usage pattern of their pitches in a significant way. For this exercise, I focused on qualified starting pitchers and tracked the change in their pitch usage via FanGraphs' pitch-type data. To track those alterations, I compared starting pitchers' usage in 2013 versus their usage in 2014. Where did they try to find a new advantage? Did this attempt work?

Below are a handful of starting pitchers who changed their offerings in a big way in 2014. In addition to noting the changes, we'll try our best to come to a verdict of the effectiveness of this change.

Felix Hernandez

Alteration: more changeups, fewer fastballs

The King is yet again a Cy Young nominee in the American League, something we've been saying repeatedly for some time now. He's likely one of the three best pitchers in baseball. So, you might be wondering, why does a guy who's this good change his offerings? Felix hasn't returned my text asking him this question yet (he's a busy guy), but my guess is that he did it simply because his changeup (below) is perhaps the most devastating of its classification in baseball. If you have that kind of changeup at your disposal, why not go to it? Interestingly, the reliance on changeups resulted in an increased value in Felix's fastball, which also saw its velocity increase by .5 mph in 2014.

Verdict: for the better

Doug Fister

Alteration: more fastballs, fewer curveballs

This is a great case for not using ERA to evaluate a pitcher, as Doug Fister struggled in his first season as a National. His FIP was over 1.5 runs higher than his ERA over his 164 innings in 2014, a season in which his fastball velocity declined, his slider remained ineffective and his curveball (below) dropped below average. It makes sense that Fister would have gone away from the curve, and while he went to the heat more frequently, he could have also considered using his changeup more, as it was his best pitch in 2014. Instead, he used it 2% less frequently than he did in Detroit and used his fastball 13% more often to account for the lack of curveballs. Battling injuries can't have helped his cause, so I'm not sure this story is over.

Verdict: mattered little

Mark Buehrle

Alteration: more curveballs, fewer cutters

Buehrle is fun because he throws a proverbial wrench in the spokes of what we often think we know about pitchers. He'll turn 36 next season, was drafted 1139th overall in the 1998 draft, and had a average fastball velocity under 84 mph last season, yet he tied for 16th in fWAR in 2014 among qualified starters. As Buehrle's velocity continues to decline, he moved away from the cutter (below), his second-hardest pitch, largely in favor of more curveballs this season. His cutter usage was nearly cut in half (from 27.4% in 2013 to 15.5% in 2014) while he upped his curveball rate 5.8% (to 13% total). The rest of the difference was somewhat evenly distributed among his other offerings, with the changeup getting 2.6% more use and his fastball being used 4.5% more often. None of Buehrle's pitches rate well from a value standpoint, so there wasn't a big gain here.

Verdict: mattered little

Jorge de la Rosa

Alteration: more cutters (kinda), fewer sliders (kinda)

Okay, I'll admit that this is one part shenanigans and one part velocity gain. de la Rosa has notably battled injuries over the last several seasons, and he missed his career high in innings by only 2/3 of an inning this season. It's no surprise that a healthy pitcher throws harder than an injured one; sure enough, de la Rosa's velo was up across the board. His fastball ticked up 1.2 mph and his breaking stuff was firmer, too. This firmness may have led to some his sliders (below) from last year (84.6 mph) being classified as cutters (87 mph) this year. Of course, there's also been discussion of de la Rosa adding a cutter this season to mix and match with his changeup, which proved to be a very effective combo. Considering he didn't throw a single cutter over the last three seasons, 2014 was a big change. We can wonder whether all of the pitch classifications are accurate, since cutters and sliders are often easy to confuse. No matter the intent, it's clear that de la Rosa added a cutter to his repertoire, although it was a below-average pitch and his slider rates considerably better.

Verdict: for the better

Zack Greinke

Alteration: more sliders (kinda), fewer cutters (kinda)

Looks like we have the inverse situation here with Greinke that we saw with de la Rosa above. While the latter added some velo and subtracted vertical break from his slider to create a cutter, the former did just the opposite, taking 1.5 mph off his cutter and added some depth to the pitch, resulting in it being classified as a slider (below). His 17.1% cutter rate in 2013 (with 2.2% sliders) turned into a 19.4% slider rate in 2014 (with zero cutters). In essence, he altered the pitch but didn't change the frequency with which he used it. And this is where we reach the fringes of pitch classification and the usefulness of labels. Does Greinke say he ditched the cutter for a slider of does he simply say that he altered his grip just slightly to generate more horizontal break and reduce the velocity of the pitch? Like Felix, he hasn't texted me back (he's a busy guy, too).

Verdict: mattered little

CJ Wilson

Alteration: more changeups, fewer sliders

In case you missed it, CJ Wilson was kinda terrible this year. The culprit seemed to be walks, which he issued with the highest frequency he's ever allowed since transitioning to the rotation back in 2010. His use of the slider (below) dipped sharply, almost 9%, and he used the changeup rose from 5.9% to 12.5% on the season. Unfortunately, the changeup rated as a below-average pitch while Wilson's slider was easily his most valuable offering. He generated nearly 4% fewer swings in 2014 with the altered offering slate, which surely didn't help his cause. Will he go back to the slider in next year? It would appear that he should certainly use it more than he did this year.

Verdict: for the worse

Jon Lester

Alteration: more cutters, more curveballs, (way) fewer changeups

Arguably the top free agent pitcher available in this year's class, Jon Lester altered his repertoire significantly in 2014. With a long track record of success, Lester rolled the dice and went away from the changeup that he'd used frequently in the past to focus on cutters (below) and curves. The move paid off. His changeup rate disintegrated from 12.5% in 2013 to just 2.6% in 2014. He shifted the usage to the cutter (23.2% in 2013, 30.8% in 2014) and, to a lesser extent, the curveball (11.9% in 2013, 16.4% in 2014). The seldom-used change was an average pitch, but the cutter took a step forward to become better than average and the curve improved significantly, proving itself far above average. I wouldn't expect the changeup to reappear in a significant way any time soon.

Verdict: for the better

Chris Sale

Alteration: more changeups, fewer sliders

Everyone's familiar with Chris Sale's delivery, which many credit with a large chunk of his success. But it's unorthodox and there is reason for concern of Sale's health given the alleged strain his delivery places on his elbow. The slider, a pitch that's been attributed to injuries, has been a steady part of Sale's diet; when combined with his delivery, it has thrown up some red flags. In 2014, he dropped his usage of the slider (below) from a staggering 29.6% in 2013 to "just" 18.4% in 2014. He pushed the usage over to his changeup, increasing it from 19% to 28.6% over the same time frame. And here's the thing: all of Sale's pitches rate well above average, so even with the swap, he still hammered hitters all year long. Chris Sale is filthy, period.

Verdict: mattered little

Some of these trends will hold, some of them will not. Pitchers are always evolving, always looking for an edge as this game continues to sway. If we look close enough, there's always something going on under the surface, as is evidenced above. The aforementioned pitchers are simply microcosms of the attempt to find the perfect balance.

All pitch data via FanGraphs

...

Jeff Wiser is a featured writer at Beyond the Box Score and co-author of Inside the 'Zona, an analytical look at the Arizona Diamondbacks. You can find his work on craft beer at BeerGraphs and follow him on Twitter @OutfieldGrass24.