How Masahiro Tanaka Bears Down

Author’s note: literally simultaneous to the publishing of this post, news emerged that Tanaka is getting an MRI on his arm. Welp! Nothing below is entirely invalidated, but it’s been a while since I’ve had such lousy timing. Let us all learn a valuable lesson.


When we analyze new players, there’s always a multi-step process. Analysis is nothing but a series of questions, and the first question is always the simplest: is the new player good? Overall, is the new player mediocre, average, good, or great? From there, over time, we start to get a little more detailed. You begin with finding out whether a player is good. Then you can start to understand how a player is good. Within the majors, there are plenty of good players, but they all achieve that level in their own particular ways, and it’s that variety that helps to keep this analysis fresh.

It didn’t take long to answer the first question about Masahiro Tanaka. Yeah, he’s good. He’s real good, as a matter of fact, and he’s one of the players most responsible for keeping the Yankees somewhere within the race. How does Tanaka succeed? Well, we all saw the splitter coming, and, yep, he throws a dynamite splitter, and with his command and unhittability, it’s hard to imagine Tanaka not being successful. By this point we know a lot about the Yankees’ new ace, but we are still filling in details. And here’s another one: Tanaka bears down with runners in scoring position. When situations are at their most dangerous, Tanaka has responded, in an interesting fashion.

Usually we don’t mess around with situational splits. Most often, you’ll hear about players and teams hitting with runners in scoring position, but the thing about hitting is it’s reactive, and over a season everything tends to even out since hitters have limited control and they tend to face a representative population of arms. Pitching, though, isn’t nearly so reactive, as it’s pitchers who control the process. So it makes sense that pitchers could work differently in different situations, and, let’s just get into a few numbers now so you can see where I’m going with this.

With runners in scoring position on the year, Tanaka has allowed a .188 wOBA. That’s second-lowest in baseball, a point behind Adam Wainwright, and Tanaka has pitched in Yankee Stadium and in the American League. With the bases empty, Tanaka has allowed a .311 wOBA. While Tanaka’s allowed 15 home runs, a dozen of those have been solo shots, and only one has come with runners in scoring position. There’s reason here to dig further.

At this point, 50 different pitchers have thrown at least 25 innings with bases empty and with runners in scoring position. Among them, Tanaka has shown the biggest increase in strikeout rate, at +10 percentage points. He’s shown the biggest increase in groundball rate, at +15 percentage points. His walk rate has hardly budged, so his K%-BB% has been markedly better, and Tanaka’s one of just five pitchers who’ve dropped their wOBAs by at least 100 points with runners on second and/or third. The message is this: in threatening situations, Tanaka has been extraordinary successful.

This might explain why Tanaka’s ERA is more than a half run lower than his FIP. This might explain why, so far, Tanaka has posted the highest strand rate in the AL, higher even than the miracle that is 2014 Chris Young. So the first thing we see is, Tanaka’s splits are fascinating. Then you have to wonder, has he pitched differently, or is this just kind of random noise? Sure enough, these splits are supported by an altered process.

Per usual, data has been drawn from both Brooks Baseball and Baseball Savant. Maybe the easiest way to understand this is by looking at Tanaka’s various rates of splitters thrown. The splitter, of course, is his specialty pitch, his putaway pitch, the pitch that might be one of the five or ten best in the world. So how has Masahiro Tanaka used his splitter, situationally?

Bases empty: 21% splitters
Scoring position: 38%

Masahiro Tanaka’s splitter has yielded one home run, and it was to the very first batter he faced in the season. Between the two situations above, Tanaka has maintained his slider rate, so these splitters have mostly come at the expense of fastballs. His splitter rate has been up across the board. With runners in scoring position, he’s thrown more splitters for the first pitch. He’s thrown more when behind in the count, and he’s thrown more when ahead in the count, and he’s thrown more when even. In those threatening situations, Tanaka has thrown 60% splitters with two strikes. Never is Tanaka’s splitter entirely absent, but with runners in scoring position, it’s been his primary pitch.

We can keep looking at more data, most of which follows. Prepare for a bunch of simple splits.

Bases empty: 37% zone rate
Scoring position: 30%

The splitter is usually an out-of-zone pitch. So, unsurprisingly, with runners in scoring position, Tanaka has thrown a lower rate of pitches in the strike zone.

Bases empty: 46% low pitches
Scoring position: 56%

The splitter is usually a low pitch. So, unsurprisingly, with runners in scoring position, Tanaka has thrown a higher rate of pitches at or below two feet off the ground.

Bases empty: 22% high strikes
Scoring position: 14%

Much of the real damage against Tanaka has come against strikes somewhat elevated within the zone. Unsurprisingly, from the above, with runners in scoring position, Tanaka has thrown a lower rate of pitches in the upper two-thirds of the zone.

Bases empty: 49% swing rate
Scoring position: 55%

This is critical. League-wide, batters swing more often with runners in scoring position, perhaps because they feel a little more pressure, and probably more because there’s more to be gained from a swing than there is with the bases empty. So pitchers know that, with runners in scoring position, hitters are inclined to be a little more aggressive, and Tanaka has generated a higher swing rate despite throwing a lower rate of pitches in the zone. A big part of this: it’s really hard to not swing at a good splitter. Splitters generate high swing rates, even though they generate low zone rates.

Bases empty: 71% contact
Scoring position: 58%

What happens when you get more swings at more balls, and when you increase your rate of splitters thrown? Tanaka just hasn’t allowed much contact with runners in scoring position. So this much should follow: with the bases empty, Tanaka has yielded 36 balls in play of at least 300 feet. With runners in scoring position, that number drops to four. There’s been little contact, and there’s been weak contact. There’s been a lot more groundball contact, a result that follows from the increase in low pitches.

It’s worth noting that everyone is adjusting to everyone. Tanaka is learning hitters and hitters are learning Tanaka, and since the season’s first month, hitters have reduced their scoring-position strikeouts. This while Tanaka hasn’t reduced his scoring-position splitters. So maybe that’s going to prove to be an interesting trend, but for now, I think it’s wisest to combine all the data, adding April to May and June and early July. If it turns out that it took a month for the league to adjust, we’ll come back to look at that later, but at the very least, you can’t really argue with the differences in pitcher approach.

And why might Tanaka pitch like this? It could be all about pacing and preservation. With the bases empty, there’s a lesser risk, where the worst possible result is a solo home run. Tanaka, of course, doesn’t want to give up solo home runs, but if you figure it can be somewhat damaging to throw a bunch of splitters and sliders, then it makes sense to somewhat conserve those pitches for when they matter the most. Tanaka, in theory, could throw 38% splitters in all situations, but then he might not last as long. This could be a kind of compromise, and if the end result is that Tanaka allows a few too many solo dingers, I have to imagine that’s preferable to arm issues or reduced pitch counts. Starters have to go easy on their arms, and Tanaka’s probably not the only pitcher who thinks this way. He’s just the only pitcher with Masahiro Tanaka’s splitter.

Presumably, over the long haul, Tanaka’s performance splits will end up a bit less pronounced. Where they stand now certainly feel unsustainable. But if this is Tanaka’s real process, there’s excellent reason to believe he might always strand a few more runners than usual, and that’s only going to make him even better at run prevention than you’d already guess. Masahiro Tanaka has been fascinating since even before he ever pitched to major-league opponents, but rare is the pitcher who doesn’t become markedly less fascinating over time. Eventually I’m sure we’ll all come to take Tanaka for granted, but 3+ months in, we’re all still learning.

Print This Post

Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

20 Responses to “How Masahiro Tanaka Bears Down”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Patrick says:

    Yikes, timely article given the MRI news…

    Vote -1 Vote +1

  2. vince says:

    Can’t help but thinking at how this may ultimately end up being an exercise in SSS.

    Vote -1 Vote +1

    • Definitely possible, but somewhat less likely given the difference in process. So it’s not just about the statistics. Granted, things could change now if Tanaka has an injured arm…

      Vote -1 Vote +1

  3. 21_22 says:

    i dont necessarily think the injury prevention conclusion is the most likely. couldnt throwing more splitters in more important situations really be due to only wanting to show hitters his best pitch in situations for which it is best suited?

    Vote -1 Vote +1

  4. GreenMountainBoy says:

    Tanaka’s injury was highly predictable and not surprising. His declining K/9 rate over the past 3-4 years, plus the sheer amount of pitches he’s thrown over that time, pointed to this, and although I would have liked to have him on my FBB team over the first half, I ranked him lower because I knew this was coming. Did I think it would be this early in his MLB career? No. I predicted it would happen by the end of 2015. But I’m not surprised either. Next news? TJS.

    -17 Vote -1 Vote +1

    • Paul says:

      You’re annoying.

      +5 Vote -1 Vote +1

    • Coach Buttermaker says:

      I agree that his overuse in Japan coupled with his adjusted pitching schedule with NYY was going to lead to injury problems. I drafted him with intentions of riding a big 2014 and selling high this winter. Unfortunately, it looks like the second half of my equation might fall apart.

      Vote -1 Vote +1

    • GMH says:

      You are so brilliant! Now tell me when the following pitcher will suffer a major injury:

      Year GS IP K/9
      1 37 263 6.78
      2 35 268 6.68
      3 38 280 6.64

      This pitcher averaged over 100 pitches per start every year, with 19 starts in which he threw more than 120 pitches, and twice he threw more than 140 pitches in a game.

      So when does this guy blow out his arm? Year 4? Year 5?

      Vote -1 Vote +1

      • ragnar32 says:

        Threw me for a loop there GMH before I realized season 3 includes postseason starts and IP :)

        Vote -1 Vote +1

      • Emcee Peepants says:

        Um, so who is it?

        Vote -1 Vote +1

        • Jack Glasscock's Cup says:

          MC P, this guy had 3 seasons before the seasons in question of more than 200 innings and 12 (!) afterwards; actually 13 afterwards if you include postseason. You’ll see him in Cooperstown this year.

          Vote -1 Vote +1

        • GMH says:

          Greg Maddux. Year 3 was his 1993 season, which included three postseason starts.

          He was pretty good after that season. But more importantly, his arm held up quite fine.

          Vote -1 Vote +1

      • GMH says:

        I can play the same game with Nolan Ryan (postseason innings included):

        1978 234.2 innings, 9.97 K/9
        1979 229.2 innings, 9.01 K/9
        1980 247 innings, 7.70 K/9

        Okay, Ryan’s elbow finally gave out when he was 46 in 1993, but only after another 2,400 innings, three more no hitters, two more ERA crowns, four more strikeout crowns, and the two highest K/9 rates of his career. In fact, if you measure Ryan’s performance using FIP, Ryan was better after 1980 than he was before.

        Or how about Steve Carlton (again, including postseason innings:

        1976 259.2 IP 6.95 K/9
        1977 294.2 IP 6.30 K/9
        1978 256.1 IP 5.86 K/9

        Carlton went on the best five-year run of his career after 1978, winning two Cy Young Awards, three strikeout crowns, pitched the most innings three times (including 331 IP in 1980 including postseason), and twice led in FIP. His lowest K/9 rate in the next five years was 7.6. And again, no arm problems during the next five years.

        I plead guilty to cherry picking three of the most durable pitchers of all time, but all three of these pitchers had far greater work loads than Tanaka has ever had and pitched against tougher competition than Tanaka had until this year. A declining K rate could be a sign for concern, but it could also be simply a reflection of a pitcher changing his approach, or a league of hitters that collectively gets better, even if short lived, at putting balls into play.

        Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>