Masahiro Tanaka figures to pitch better in the second half

In one sense, the Yankees find themselves in great shape at the beginning of the second half. They own a 62-34 record and have all but secured themselves a playoff spot. In another sense, however, the Bombers are in some trouble. Much like the start of the year, the Boston Red Sox have decided to stop losing again, extending their lead in the division race to 5.5 games.

The Yankees have to make up some ground if they want to avoid the Wild Card Game. To do so, they need quality starting pitching. Apart from Luis Severino and CC Sabathia, however, the current rotation isn’t getting it done. The current options available on the trading block are also uninspiring, to say the least. Should the market remain as unappealing as it is now, the Yankees’ best option might be to stick with what they have and hope for better performances.

Luckily for the Yankees, they have one pitcher who could be primed for a second-half surge. His name is Masahiro Tanaka, and the data suggests that he’s been more unlucky than bad in the first half of the season. In fact, all Tanaka needs is his HR/9 to normalize in order for his results to improve substantially.

By now, it’s well known that Tanaka has a home run problem. After running a HR/9 of 1.77 over 178.1 innings in 2017, he has reached new heights (depths?) in 2018, recording an astronomical 1.94 HR/9 over 83.1 innings. Over those two years, Tanaka has allowed a whopping 53 homers, the highest total in the major leagues. As a wise man once said: “It’s not what you want.”

To say that Tanaka has been plagued by the long ball all throughout his career, however, would be a misstatement. While Tanaka’s career HR/9 of 1.38 does indeed paint a picture of a homer-prone hurler, when you divide his career into individual seasons a different picture emerges. In 2014 and 2016, Tanaka’s HR/9 was actually below one. That’s more than two-fifths of his stateside career. Granted, this doesn’t disprove the notion that Tanaka is homer-prone, but it shows that Tanaka’s HR/9, as with any pitcher, is subject to substantial fluctuation.

What this suggests is that it’s probably too early to treat Tanaka’s pronounced homer problem over the last two years as a “new normal”. His current homer pace is unsustainable even for Tanaka, and like any other pitcher we can expect it to regress towards his career norms. In fact, we saw this happen just last year, during Tanaka’s second-half surge.

During the first half of 2017, Tanaka posted a HR/9 of 2.03. Down the stretch, that figure dropped to 1.41, a smidge above his career mark of 1.38. What’s interesting is that this wasn’t accompanied by any meaningful changes in his contact profile. Tanaka’s flyball rate only changed by two points, from 33.4% in the first half to 31.3% in the second half.

Meanwhile, his Hard% went from 32.1% in the first half to 30.4% in the second half. Both improved, but not by much, and certainly not enough to warrant such a drastic reduction in HR/9. Tanaka just got less unlucky on a few more fly balls than he did in the first half, and that made a substantial difference on the trajectory of his season.

Indeed, what would happen if one substituted Tanaka’s current 1.94 HR/9 with his career average of 1.38? That would generate approximately 13 homers allowed in 83.1 innings, five fewer than the 18 he’s already given up. Even if one assumes that the subtracted homers were all solo shots, five less homers would bring Tanaka’s ERA down to 4.00, a much more respectable number than his current 4.54 mark.

Granted, Tanaka has a couple of problems beyond his homers. His walk rate is trending in the wrong direction, and he’s giving up fly balls at a career-high rate. Tanaka’s overall profile remains strong, however, and he still has excellent stuff (13.8% whiff rate, 8th in MLB among pitchers with more than 80 innings pitched). If a few more fly balls start to go his way, he’s a solid mid-rotation starter. Boy, do the Yankees ever need one of those right now.

All statistics courtesy of FanGraphs.com.