Kuroda’s four years

Last Sunday, 24 hours after the Cardinals wrapped up the 2011 season with their 11th World Series title, the major league free agency season began. Though the headlines will be dominated by reports on superstars such as Albert Pujols, Prince Fielder and Jose Reyes, there are plenty of players who, despite a lower profile, will be able to give their new team a lift. One of these players is starter Hiroki Kuroda, who has spent four seasons in the major leagues as a Los Angeles Dodger after signing out of Japan after the 2007 season.

Kuroda has contributed 699 regular season innings in his major league career. Comparing him to the 62 other pitchers with at least 100 games started since 2008 gives us some context about his performance in some standard pitching categories.

Metric  Rank
ERA      16      
K%       37
BB%      7
GB%      12

These data are courtesy of Fangraphs/Baseball Info Solutions. The metrics from here on in this post will be from Gameday stringers.

He’s done a good job at keeping the ball on the ground and limiting walks. Despite being middle of the pack in strikeouts for his career, he’s been significantly above average in his past two seasons. His groundball rate has taken quite a hit recently, so we’ll be looking at that once we get into Kuroda’s “stuff” and how he approaches hitters.

Repertoire

Using PITCHf/x data, Kuroda’s pitch arsenal can be identified as consisting of, for the most part, five pitches.

His bread-and-butter pitch has been a two-seam fastball which he’ll throw at about 92 mph. At about the same velocity, he’ll also show hitters a four-seam fastball, which sometimes has some cutting action on it.

His primary offspeed pitch his a tight slider which he’ll throw throughout the 80s. A few miles per hour harder is his split-fingered fastball, which has a little bit of action away from lefties and breaks downward more than the slider.

Kuroda has been throwing a curveball in the high-70s more often over the past two years and spent time working on it during spring training. Please note that his curveball is very similar to his slower sliders, so there is plenty of subjectivity in splitting up the two pitches. Another report out of spring training had Kuroda toying with a cutter; I haven’t found any difference in his pitches comparing 2011 to previous seasons, so for the time being I will stick with these five labels for his pitches.

One last thing about the classification of Kuroda’s pitches: A handful of pitches (15 that I tagged over the 2008-2011 data set) look like more traditional change-ups than splitters. However, Kuroda is reported (from Dylan Hernandez’s Los Angeles Times article in February) to have given up learning how to throw a change-up, so I will leave that to the side as well and exclude those pitches from this analysis.

Below is a spin deflection plot and a horizontal spin/velocity plot of Kuroda’s five pitches. The movement values are from the raw data and have not been adjusted for camera variation from ballpark to ballpark.

Results

As mentioned earlier, Kuroda is good at preventing walks and fly balls, in large part due to the sinking fastball he relies on. Dating back to 2008, here are some full-season pitch metrics.

Year     #      Swing%    Whiff%     Ball%      Called%    GB%
2008     2656   49%       20%        36%        15%        49%
2009     1797   48%       20%        36%        16%        47%
2010     3026   48%       25%        36%        16%        51%
2011     3201   48%       23%        36%        16%        44%

And split up by pitch type in the tables below:

Sinker
Year    mph     #       LHB%    RHB%    Swing%  Whiff%  Ball%  Called% GB% 
2008    92.5    1019    46%     32%     53%     9%     29%     17%     58%   
2009    92.7    923     48%     53%     50%     12%    30%     20%     49%
2010    92.2    1147    35%     42%     51%     13%    30%     19%     60%
2011    91.9    1218    36%     41%     51%     13%    30%     21%     50%
All     92.3    4307    40%     42%     51%     12%    29%     19%     55%

Fastball
Year    mph     #       LHB%    RHB%    Swing%  Whiff%  Ball%  Called% GB% 
2008    92.2    576     18%     27%     50%     17%    38%     12%     36%   
2009    92.1    223     20%     5%      53%     18%    42%     4%      28%
2010    92.0    542     25%     12%     49%     23%    42%     10%     39%
2011    92.1    713     29%     16%     46%     23%    43%     11%     29%
All     92.1    2054    24%     16%     49%     21%    41%     10%     34%

Slider
Year    mph     #       LHB%    RHB%    Swing%  Whiff%  Ball%  Called% GB% 
2008    84.1    682     20%     34%     44%     39%    37%     19%     38%   
2009    84.3    451     17%     34%     42%     33%    39%     19%     41%
2010    84.7    831     14%     40%     47%     37%    35%     18%     39%
2011    84.2    630     8%      30%     49%     33%    36%     15%     38%
All     84.3    2594    14%     35%     46%     36%    37%     18%     39%

Splitter
Year    mph     #       LHB%    RHB%    Swing%  Whiff%  Ball%  Called% GB% 
2008    88.1    281     15%     7%      50%     31%    46%     4%      53%   
2009    88.5    158     13%     5%      54%     44%    42%     4%      58%
2010    87.1    310     17%     5%      51%     38%    42%     7%      58%
2011    86.9    474     20%     10%     51%     36%    35%     4%      53%
All     87.4    1223    17%     7%      51%     37%    44%     5%      55%

Curveball
Year    mph     #       LHB%    RHB%    Swing%  Whiff%  Ball%  Called% GB% 
2008    79.6    24      2%     <1%      33%     25%    42%     25%     33%   
2009    80.6    21      2%     <1%      5%      0%     52%     43%     100%
2010    80.3    148     9%      1%      22%     30%    45%     33%     36%
2011    79.6    145     7%      2%      28%     22%    33%     39%     29%
All     87.4    338     5%      1%      25%     25%    40%     36%     34%

You can use this post as your reference for these metrics.

First off, let’s address the groundball decrease alluded to at the beginning of this post. The sinker’s groundball rate has fluctuated in the 50-60 percent range over his career, and 2010 was one of the down years. Combine this with the fact that he threw a greater percentage of four-seamers (an extreme flyball pitch) in 2011 than he did in 2009 and 2010 and you get a different looking batted-ball distribution.

But at the same time, Kuroda is benefiting from the extra four-seamers he threw last year because it is a much better strikeout pitch than his two-seamer. His 21 percent whiff rate for the pitch is significantly above the league average (16-17 percent), and it’s been on the rise over the past two years. This is probably due mostly to its cutting action relative to the average four-seam fastball; in the 2011 THT Annual, Jeremy Greenhouse found that those kinds of fastballs typically have higher whiff rates than those with average movement. So while Kuroda has sacrificed some ground balls, he is picking up more strikeouts.

By the way, in case you want to see the change in Kuroda’s fastball distributions on a game-by-game basis, I’ve provided the chart below. Notice that for his first 10 or so starts in the major leagues, he relied heavily on his four-seamer. He used it about as much as his two-seamer in August and September of 2011, and has otherwise been a pretty generic sinkerballer.

The offspeed pitches: how they’re different

Kuroda’s slider and splitter are both good strikeout pitches, as they both have whiff-per-swing rates in the high 30s. However, the splitter generates a ton of ground balls while the slider does not. There are three primary reasons for this:

Pitch movement. As you could see in the spin deflection chart earlier in this post, Kuroda’s splitter drops significantly—about 16 inches on average—more than his slider does. John Walsh in 2007 and Josh Weinstock in 2011 have shown us that vertical movement plays a significant role in predicting groundball rate. (Their studies applied to fastballs, but the same principle should generally hold true for other pitch types).

Velocity. In addition to breaking more, Kuroda’s splitter is also three mph harder than his slider. Josh Weinstock’s piece also found that velocity correlates with groundball rate; this meshes well with Josh Smolow’s research from last December.

Pitch location. Kuroda really likes to keep his splitter low; since 2008, 63 percent of his splitters have been below the bottom of the strike zone. The slider, on the other hand, catches a lot more of the plate and has been located on average 10 inches higher in the strike zone than the splitter. The vertical location/groundball correlation is something that Weinstock also looked at in his IATMS piece, and his conclusions point to a strong relationship between the two factors. The graph below shows the difference in vertical location between Kuroda’s slider and splitter.

In conclusion

There’s a lot to like about Kuroda’s game. He throws strikes and has shown the ability to get both grounders and strikeouts, though the four-seam/two-seam combo prevents him from being extreme in either category. Pitch selection is partially to blame for his groundball decrease in 2011; another year of data would be nice in helping to identify what else is going on. He appears much more prone to hanging his slider than his splitter, but gets good results with both.

It’s not all positives, though. The elephant in the room is that Kuroda is by no means young. He will be 37 in February, so I can’t imagine that any team would be willing to give him any kind of serious money for more than two years. The other thing we shouldn’t forget is that he’s pitched his whole major league career in the NL West, so a move east would probably entail some kind of decline in performance. Kuroda himself is apparently pushing to stay in Los Angeles, or perhaps return to Japan to finish his career. All in all, however, his well-rounded pitching approach will make him appealing as a middle-rotation guy for many teams this offseason if he chooses to pitch for another major league team.

References & Resources
PITCHf/x data are from MLB Advanced Media and Sportvision; they are here courtesy of Joe Lefkowitz’s PITCHf/x tool. Pitch classifications are by the author.