October 28, 2013

Playoff Prospectus

World Series Game Four Recap (Red Sox Edition)

by Sam Miller

Printer-friendly

Contact Author

Throughout the World Series, we'll be providing two recaps of each game, one with a focus on the winner and the other devoting a longer look to the loser. This is the Red Sox entry for Game Four. The Cardinals edition is here.

During the middle stages of Game Four, Ben made this point:

Forget bunts, IBBs, etc. Leaving pitchers in too long is the new thing that bothers us most about managers. Probably matters more, too.

— Ben Lindbergh (@ben_lindbergh) October 28, 2013

I'm pretty confident that Ben's going to bring that point up in his recap of this game from the Cardinals' perspective, but my piece is going up before his piece so I'm stealing it, too. Clay Buchholz pitched four innings for the Red Sox on Sunday night. That was, depending on where you happened to be in the universe's timeline, either an inning too long

90 mph with no life, against a good lineup. This might get ugly.

— keithlaw (@keithlaw) October 28, 2013

or two innings too long

I wouldn't let Buchholz hit here. Carp vs Lynn a good match-up, since you know Matheny won't get a LHP to counter.

— David Cameron (@DCameronFG) October 28, 2013

or even four innings too long:

If my starting pitcher was down 4 Mph on fastball there is no chance I am using him. Do they not have guns in the bullpen?

— Mitchel Lichtman (@mitchellichtman) October 28, 2013

Each point was compelling, but Lichtman's especially is intriguing. The first pitch Buchholz threw was 86 mph, low, with tailing movement. I scribbled down changeup. The second pitch was a two-seamer, away, at 88 mph. At that point, I realized that the first pitch was a fastball. At that point, I would have thought about pulling Buchholz.

Despite our inability to figure out which team characteristics correlate to postseason success, it is inescapable that postseason baseball is different than regular season baseball. The most obvious and, for a manager, most important difference is that one of the things that is in short supply during an extremely long, day-after-day regular season—competent and healthy pitching—becomes bountiful in October, when off-days come every 48 or 72 hours, series are divided by one to four off-days, and the certain nearness of the end makes it easier for everybody to leave it all on the field. There's nothing to save it for, anymore.

The scarcity of quality pitching guides baseball strategy at the most basic levels: It's why Justin Verlander pitches seven innings per start instead of five, and it's also why he pitches seven innings per start instead of nine. And so, if this were the regular season, you could understand why a team would turn to a right-handed starter with an 88 mph fastball to eat some innings. It's not, though, and so you could really understand why the Red Sox might have considered scratching Buchholz and asking for two or three innings from Ryan Dempster, then two or three innings from Felix Doubront. Maybe Buchholz gets an inning in there, as a reliever, and can pump his fastball up to 92 in a short burst. Then ask for four innings from Tazawa/Lackey/Uehara. You never have to let a pitcher bat. If your offense scores a bunch you let Dempster keep going, or bring in the back of your bullpen. It makes more sense than letting, effectively, Jason Marquis start Game Four of the World Series. It's worth wondering why they didn't do that.

One possibility is that, even seeing him in the bullpen, they didn't realize it was going to be as bad as it was. Buchholz's shoulder fatigue was well known; his poor results deep in games was well known; but, even in two bad postseason starts, his velocity was down only about a half mile per hour from the regular season, to about 92. Things got worse quickly, to 88 for his two-seamer and 89 for his four-seamer in this game. Worse, his changeup velocity didn't drop by as much, so the nine mph differential between it and his sinker was significantly less than the 11.5 mph differential in the regular season. His cutter, though, had slowed by as much, which may explain why he didn't throw it as often as he typically does. (It's hard to say for certain that he didn't. As Dan Brooks noted, "Gameday has no idea what Buchholz is throwing because he is ~5mph off his normal stuff. Hard to blame it." Brooks' cutter count—eight all game, down from about 25 per postseason start—is reliable, though.)

A second possibility is that, even acknowledging that velocity loss, they thought he was still capable of getting batters out. Maybe he felt sharp in the bullpen, aside from the velocity. Maybe he had good feel. His sinker was moving nine horizontal inches, on average; that's compared to five inches in the postseason before this start, and six inches in the regular season. His changeup had, on average, four inches of tailing movement; before this, it averaged no tailing movement at all. His cutter, for lack of use, did induce three whiffs in eight pitches. His changeup, for lack of use, was consistently down and didn't miss over the plate once all night. His fastball command seemed good to me. You could do worse.

And the third possibility, the most likely, is that the Red Sox knew they were playing with fire, but they also knew that, this being the postseason, they could treat Buchholz very differently than they would normally treat the guy who takes the mound in the first inning. In his previous three postseason starts, Buchholz allowed one run, combined, in the first four innings. Each time, things got out of hand quickly. But knowing they had Dempster, Doubront, Lackey, etc., they might have gone into this game with renewed will power. Knowing that they could pull Buchholz after four, they could gamble on him. They would just need the will power to hold themselves to it, and hope that anything that went wrong would go wrong slowly enough for them to back out of it.

And it worked. That's the twist to the whole thing. Buchholz never looked good, he always looked uncomfortable, he walked three and struck out two, he left two runners on base in each of the final three frames. He probably shouldn't have been out there. But he was, ultimately, the hero, or at least the gutsiest gutter who ever gutted. He gave up one hard-hit ball, or maybe three, depending on definitions. The run he earned was unearned, and maybe blamed not even on his defenders but on the football-damaged outfield. He lowered his ERA in the first four innings this October to 0.56. Sometimes, even in October, a manager just has to close his eyes and hope that the lesser player wins, and sometimes that works.

Speaking of decisions that pay off that don't seem likely to pay off: When Shane Victorino was scratched with lower back tightness, Jonny Gomes was inserted into the lineup. Lance Lynn has a massive platoon split in his career, and a repertoire and delivery that back up that split. Mike Carp hit .300/.367/.537 against right-handed pitching this year, and started 33 games in the outfield in exactly these sorts of situations. Instead, Gomes—fresh off his first postseason loss as a starter—got to play against the right-hander. He drew a very impressive walk against Lynn

and then, against the right-hander Seth Maness, hit the three-run home run that won the game. Like holding on a meaningless baserunner in the ninth, or letting an injured pitcher start in the World Series, sometimes moves just work.

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