April 11, 2014

BP Unfiltered

This Week's New Pitching Lines, 4/11

by Sam Miller

As was immediately identified by everybody with a Play Index subscription and/or common sense, Danny Salazar's 3.2/6/5/5/2/10 pitching line yesterday was a unique pitching line. Shoot, the bookends alone are brand new; all the hits, runs, and walks are irrelevant, as there has never been a 3.2 inning, 10-K start before. To this, the esteemed Zachary Levine suggested that

@SamMillerBP @Toirtap This needs to be its own installment in the series.

— Zachary Levine (@zacharylevine) April 11, 2014

But with due respect to Zachary, I would argue that Salazar's pitching line—while interesting, and deserving of attention throughout the baseball world—is the opposite of what this series lusts after. It's a freak-show start, and baseball has tons of freak shows, the sort that Jayson Stark and YCPB and High Heat Stats and others chronicle so wonderfully. What's most interesting to this series is the way that even the banal can slip past baseball's century of monkeys and typewriters. The way that Salazar's teammate, Zach McAllister, could so quietly make his own kind of history with a simple little 7.2/5/0/0/0/7 line this week. Randall Delgado's 3.1/6/3/3/3/0 yesterday has been done eight times, because everything in baseball has been done eight times, yet somehow nobody has ever produced a 7.2/5/0/0/0/7. And nobody cares. Anyway, you've now seen two of this week's brand new pitching lines. The others:

The best new line of the week: Stephen Strasburg's 6.2/3/1/1/1/12
Seems like almost every line Strasburg produces is either new or darned close. His first start this year produced a third-time-ever 6/5/4/4/2/10*. His second start produced a first-time-ever 4.1/8/6/3/3/6 line. And now we have another all-new line in his third. There are two factors at work in Strasburg's lines: He strikes out more batters than all but a handful of Ace-types; and yet his managers have never really let him go like an Ace-type. The Venn diagram between pitchers who pitch like Strasburg and pitchers who get pulled as early as Strasburg does shows very little overlap.

In this start, Strasburg was pulled after 98 pitches, with two outs in the seventh and the tying run on first. Managers hate to pull a starter mid-inning when he's doing well. Consider:

• 7 innings, 10+ strikeouts, <3 runs allowed: 969 times in history
• 6.2 innings, 10+ strikeouts, <3 runs allowed: 98 times in history

That managers have historically let the Strasburg-type finish this inning doesn't necessarily make it right, and it's sort of interesting and exciting that Matt Williams brought in Jerry Blevins. Williams has professed an openness to statistical best practices, and this move suggests he's not going to fall victim to a common managerial mistake: Trusting his famous-name starter instead of his not-famous reliever even late in the game, when the reliever is almost certainly a better option than a tiring starter going a third or fourth time through the order. Jerry Blevins came in for Strasburg and retired all three lefties lined up. It did come at some expense—a double-switch required—and, of course, one might wonder whether it communicates anything to Strasburg, and whether that matters:

Strasburg felt he could have pitched past the seventh, but Williams went to his bullpen to finish the game.

"I was definitely seeing the finish line, but you have to come out when skip goes out there and takes the ball out of your hands," Strasburg said.

So Williams' move was not obviously right (or wrong). But for a manager who promises to try new things, it's encouraging that he tried something that is literally brand new.

*Gerrit Cole very nearly matched this line yesterday, but for an unearned run that turned his line into a new-line masterpiece: 6/5/4/3/2/10. It took 100 years for baseball to produce the perfect odometer reading.

The worst new line of the week: Hyun-Jin Ryu's 2/8/8/6/3/2
The unearned runs make it much less likely to be repeated any time soon, but even if those runs were earned his line would be unique. Just as it's a surprise to see a great pitching line that includes a partial inning, it's a surprise to see a terrible pitching line that doesn't include a partial inning. If Ryu had been left in to get one more out, then the line (other than the unearned runs) wouldn't have been new. If he'd been left in for two more outs, he would have been just the sixth pitcher to get a 2.2/8/8/6/3/2. It's hard to be so bad that you get run out of an inning this early without even getting one out. (Jordan Zimmermann's 1.2/7/5/5/1/2, for instance, was a fourth-timer.) Ryu's start is the lowest Game Score of the year, incidentally.

The most surprisingly new line of the week: Masahiro Tanaka's 7/7/3/3/1/10
This is what we live for: A brand new line that involves neither a partial inning, an unearned run, or some outside force (injury, rain) that unnaturally ends an outing. More than that is this paragraph I wrote two years ago when I first explored baseball's database of pitching lines:

So which previously unpitched lines are you most likely to see when you go to a game? There are a few that look like they should have been done by now. There’s 6/4/4/4/0/4, and 6/7/0/0/2/7, and 6/7/1/1/3/7, and 7/7/3/3/1/10, and 7/10/1/1/0/3.

7/7/3/3/1/10! I feel so alive! I feel so alive for the very first time. And I think I can fly.

The rest:

• Jordan Zimmermann’s 5/4/1/1/1/9
• Marco Estrada’s 5.2/4/2/1/3/6
• Bruce Chen’s 6.1/6/1/0/0/7
• Cliff Lee's 7/10/0/0/0/6
• Madison Bumgarner's 6.1/8/2/2/1/10
• Clay Buchholz's 4.1/13/6/6/0/3
• Jon Lester's 7.1/7/4/2/1/6
• Jered Weaver's 5.2/5/5/5/06
• A.J. Burnett's 5.2/5/8/4/6/3
• Ubaldo Jimenez's 4.2/8/4/4/5/4
• Tyson Ross' 5.1/5/7/2/5/2
• Edwin Jackson's 4.2/9/6/6/4/4
• Robbie Ross' 5.1/2/1/0/6/4
• Roberto Hernandez's 5/7/4/3/1/9
• Lucas Harrell's 4.1/7/5/4/3/2
• Anibal Sanchez's 56/6/3/1/1/6

Sam Miller is an author of Baseball Prospectus. Follow @SamMillerBB
Click here to see Sam's other articles. You can contact Sam by clicking here