Mark J. Rebilas-US PRESSWIRE
A PITCHf/x look at Yu Darvish's rookie year, and how it compared to the rest of the league.
As we hit the doldrums of the MLB off-season, we are short on interesting topics for the Texas Rangers at the time being. The future is kind of nebulous, and most of the options have already been covered; it is just a matter of waiting and seeing who Texas installs in to their 2013 roster. The recent past is not particularly fun to continue to rehash, either.
So let's talk about Yu Darvish!
There has now been one full season of Yu Darvish pitching in American League parks, in front of PITCHF/x cameras, so we can now take a look at how his repertoire shapes up beyond just eye test and anecdote. The results are actually pretty fun.
Like most pitchers, Darvish's most frequently-thrown pitch was his four-seam fastball. Typically, of course, establishing your basic fastball for strikes to get ahead in the count allows you force hitters to go after secondary pitches, and Darvish's approach was no different. In fact, even with his vast array of pitches, he threw about as many heaters as the average MLB starter: 31.5% of the time for Darvish, against 32.9% for MLB.
That fastball came in about a tick higher than MLB average, sitting just under 93 miles per hour, and topping 97. What was not average was the amount of horizontal movement Darvish got, with his four seamer coming in nearly two inches further on righties than the average MLB pitcher's. That movement may have been a bit of a curse through the middle part of the year, where Darvish struggled to get ahead in counts. A shifter fastball must certainly be harder to throw for a strike, and until he figured out where to put that pitch at the start of at bats, he struggled to keep his walk totals down. His four seamer also failed to garner more than a slightly-above-average 8.9% swinging strike rate.
It gets more fun from here, though. His two-seam fastball actually resulted in a swinging strike just 8.1% of the time, but it made up for it elsewhere. The average MLB pitcher used a two-seamer 14% of the time in 2012. Darvish pulled it out at least 17.3% of the time.
The other fastball was nearly as quick as the standard selection, which made it a serious weapon compared to the rest of the league, where it averages 90 mph. Not only was their a rocket engine behind the pitch, but it moved ridiculously. While the average MLB two-seamer was moving just under an inch horizontally, Darvish was moving his two seamer more than eight inches, as well as dropping it around an inch more than MLB average. Remember when we all thought he might be throwing a shuuto?
That ridiculous pitch wasn't getting a lot of swing and misses, but it did get a ground ball more than half the time contact was made (52.4% to be exact). While his other delicious secondary pitches were punching people out, his two-seam was getting people to roll over pitches in to harmless in play outs.
Every pitch after those was completely nasty. The Major Leagues don't see a lot of splitters anymore, with starters throwing them just just over once in every 100 pitches. Yu Darvish, however, busted his out around once in every twenty offerings, and it was awesome when he did. The splitter was the first in a trio of pitches getting hitters to whiff (often hilariously) just over 11% of the time, and fairly strong measure of pitch effectiveness.
Also getting a whiff more than 11% of the time (actually closer to 12%) was the cutter. Once again, Darvish used a secondary pitch more than the rest of baseball, throwing cutters 17% of the time, compared to an average of 6%. Similar to the splitter, when (rarely) hitters did manage to connect, the result was more often than not a ground ball (55% of the time), resulting in a BABIP of just over .200.
That total is half of what the other breaking pitch accomplished, though. Darvish's slider, at just over 18%, registered "only" about five percentage points ahead of MLB selection average. It was his second most used pitch, though, and for good reason. At 81.6 against an 83.5 average, his slider was actually relatively slow (by recorded data). Of course, it moved more. A lot more. With more than four times the average horizontal break, at 8.1 inches, and with more sink to boot.