Baseball Prospectus is looking for a Public Data Services Director. Read the description here.

This series has been compelling from the start, but it took until Game Five for it to look like a contest between two of the best teams in baseball. Game Five was the first without an error. It was mercifully free of egregiously bad baserunning, and it didn’t end with a debatable call. With the memory of Dana DeMuth's floating strike zone still fresh, it felt well-officiated behind home plate, aside from this third-inning strike three to Matt Carpenter.

And so it came down to a duel between two talented pitchers. Adam Wainwright, who called his Game One delivery “horrible,” made some “great adjustments” and “executed [his] plan all night long.” Lester, who excelled in Game One, was even more efficient in Game Five, again getting 23 outs and finishing with an almost identical line, but this time throwing 21 fewer pitches.

Because it was so close, we can point to individual pitches that proved pivotal, like the 0-2 curveball that Wainwright left a little up to Dustin Pedroia in the first

or the 1-0 four-seamer Matt Holliday drove out to center in the fourth.

But the most pivotal pitches came in the seventh. By then, Wainwright may have begun to tire. The Cardinals swung at fewer first pitches from Lester, but the Red Sox, as they tend to do, made the pitcher opposing them work harder. “I wanted to attack them today, get them into my pitcher’s counts, keep them out of those deep counts, where they get your pitch count up early,” Wainwright said. And for the most part, he accomplished those goals—just not quite as well as Lester did during the home halves. Cardinals batters saw 3.50 pitches per plate appearance; Red Sox batters saw 3.82. And so through six innings, Lester had thrown only 69 pitches, while Wainwright had thrown 86.

Maybe that made the difference. In the seventh, Wainwright lost a little command. With one out and Xander Bogaerts on first, he got ahead of Stephen Drew 1-2 and then threw three straight curves out of the strike zone, later acknowledging that he’d “spun out of [his] delivery” on the last two. And the next batter, David Ross, put the game away with a ground rule double. A foot to the left, and it would’ve been foul; a foot to the right, and it might have scored two instead of bouncing into the stands.

After Ross’ first at-bat, Gabe Kapler noted that Wainwright had him set up for a curve in his second trip to the plate:

But when Ross singled in his second at-bat, he still hadn’t seen the curve. It wasn’t until that seventh-inning matchup, when Wainwright really needed an out, that he finally broke out the breaking ball. The 1-2 curve missed its target, but it was low enough to work anyway against a hitter who hadn’t seen a similar pitch since Game One. Unfortunately for Wainwright—and despite all the groundwork the right-hander had laid—Ross was ready.

(As for why Wainwright was still in in the seventh—and why he stayed in to allow another run to the left-handed Jacoby Ellsbury, whom he was facing for the fourth time—see my story from Monday about Mike Matheny’s slow hook. I said then that I was starting to sound like a broken record; now my needle is officially stuck.)


The first scary thing about the performance of David Ortiz, who’s hitting .733/.750/1.267 on the series, is that his slash line would have been even better if Carlos Beltran hadn’t robbed him of a homer in Game One. The second scary thing is that the Cardinals have probably pitched him the way they wanted to. Here’s the location plot vs. St. Louis:

That bright red square where Ortiz has seen the highest concentration of pitches is the same as the one where he saw the highest concentration of pitches during the regular season. It’s low and away, where pitchers can avoid his premium power zone (low and inside) but still hope to earn the outside strike that’s often called on left-handed hitters. So the Cardinals have followed the same strategy employed by Ortiz’ previous opponents; it just hasn’t worked as well for them (not that it’s proved foolproof for anyone).

According to Brooks Baseball, Ortiz has seen about the same distribution of fastballs, breaking balls, and changeups in this series that he did during the regular season. (He’s batting .900 and slugging 1.400 on the hard stuff, and.500/1.500 on off-speed offerings, with only curves and sliders meeting with anything close to success.) He’s swinging about as often as usual—in Game Five, he uncharacteristically offered at the first pitch three out of four times, perhaps figuring that he might not have another chance. But when he’s swung, he’s missed much less often:

Pitch type

Regular Season Whiff/Swing %

World Series Whiff/Swing %










To some extent, Ortiz has benefited from perfect placement of his batted balls; his grounder in the eighth inning last night probably would have been fielded if the Cardinals hadn’t been guarding the lines (or if Allen Craig could move), and his infield single in the second probably would have wound up in a middle infielder’s glove if it hadn’t hit Wainwright. And as Sam Miller and I discussed on the podcast today, even if Ortiz was particularly locked in from Games One through Five, his hot hitting won’t necessarily extend to Game Six. If the Cardinals get away from their game plan—either by varying their own timing or position on the rubber when Ortiz is up, as Wainwright did, or by not making a move when a lefty reliever is called for because of small-sample stats—they’re as likely to hurt themselves as they are their new nemesis.


When I said Game Five was the first without an error, I meant the kind decided by the official scorer. Managerial mistakes are another matter. I won’t call John Farrell’s decision not to pinch hit for Lester with one out and men on second and third in the seventh inexplicable, because it’s easily explained: Lester had thrown only 69 pitches and allowed only one run, and he was coming off a 1-2-3 bottom of the sixth. He’s Boston’s best starter, and he throws with his left hand, an advantage against a team that has struggled against southpaws. On top of that, we know that Farrell is understandably wary of going to anyone in his bullpen but Koji Uehara—hence John Lackey's cameo as a setup man—and if he’d removed Lester, he would’ve had to ask his other relievers for four or five outs.

But the fact that we can follow Farrell’s thought process—and that many, if most managers, would have made the same call—doesn’t mean that this wasn’t the worst managerial move we’ve seen since the last time Farrell let a poor-hitting pitcher make an almost automatic out instead of demothballing Mike Napoli. Lester was working through the Cardinal lineup for the third time, so the times-through-the-order penalty was in full effect, and as good as he’d been, succeeding through six isn’t predictive of success in the seventh. Even if Farrell had known that Lester would give him another 1 2/3 scoreless, the difference between that and whatever he would’ve expected his best available reliever to do over the same span couldn’t have equaled the gap in expected scoring between Lester and Napoli with two runners in scoring position. If you want more on this move, Mitchel Lichtman did the math and wrote almost 3,000 words about it, concluding that it cost the Sox roughly three percentage points of win expectancy. That doesn’t sound like a lot, but it’s about as big a bullet in the foot as a manager can insert with a single decision.

  • If you’re tired of talking about slow hooks for starters and need something new to nitpick, there’s always that old standby, batting order. Both managers did some debatable tinkering before Game Five; when the dust cleared, Shane Robinson was batting second for St. Louis and Jonny Gomes was in the cleanup slot for the Red Sox. Those weren't the greatest looks for baseball's best-hitting teams.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
Farrell said after the game that Uehara could only give him 4 outs (and based upon what I have seen in his press conferences, Farrell seems like a straight talker). So if he takes Lester out, he has to get 5 outs from Tazawa...who has pitched in 4 straight games (yes...there was a day off between 2 and 3). He cannot feel confident in any other pitcher he has (except Doubront who is not available). Both Tazawa and Uehara has publicly stated that they are tired.

Farrell made the right call.

You can argue that he took away the chance for a big inning (which would let him feel comfortable using other relievers) and you might even argue that if he takes out Lester with a low pitch count, he becomes more available for work in a possible game 7. But calling this managerial move bad journalism.

Win probabilities are great at showing how small decisions can affect the chances to win aggregate. If you are a Sox fan, you are happy that Farrell managed to win a specific game with a specific set of circumstances and left the chart at home.
Calling this managerial move bad is bad... JOURNALISM?

Either that's way too harsh or you're not using that word properly. You make some fine points in your post, but Ben's argument is perfectly valid (i.e., leaving Lester to pitch one more inning must be weighed against the decision to let him bat in the 7th in an extremely high-leverage situation; what's more, as Mitchel Lichtman has shown, even a mid-level reliever is usually preferable to a good pitcher pitching a good game against the third time through the order). Perhaps you are right that Farrell should have left the charts at home, but it's also the case that you can't solely judge decisions by results.
Except that Lichtman's point presumes you have an even mid-level reliever available to pitch those outs. The Sox don't have any reliever other than Tazawa and Uehara that fits that description right now. Breslow is completely shell-shocked and can't throw a strike to save his life. Same for Morales. Doubront was unavailable as he was probably going to be saved for rescuing Lackey if he sucks tomorrow or starting game 7. So who's left? Dempster?
"Other than Tazawa and Uehara" is a pretty big thing to brush aside, no? If I'm a Sox fan I'd rather have those guys in - or even Dempster or Workman - to pitch an extra inning with an extra run or two, than have Lester in with less wiggle room. That said, my big problem with Shawn's post is that he called Ben's perfectly logical article bad journalism, which seems to under-appreciate the pros and cons of Farrell's decision.
There is zero chance that Tazawa and Uehara could have gotten 9 outs that night. Tazawa had pitched in every game in the series up to that point, and Uehara is almost on fumes, by his own admission.
He made the right call. If Napoli pinch hit and then struck out and say Ellsbury flies out, people would be questioning why would you take a pitcher out of a game that he was dominating.
I'm not sure how this hypothetical supports your point. Couldn't you just as easily say that if Napoli pinch hits with a bases-clearing double then no one is questioning Farrell's decision? I'm not saying that's what would have happened, but I'm not sure why your series of hypotheticals (Napoli striking out, Ellsbury flying out, a reliever coming in and presumably pitching worse than Lester would have) have any weight to them.
And if I made an all in call before the flop on the first hand of the World Series of Poker with pocket aces and my opponent had kings and spiked a third king, people would be questioning why I'd risk my tournament life that early. They'd be wrong.

There's an argument that Ferrell was right to leave in Lester (check out @mitchellichtman discussion with @tangotiger on twitter last night), but it isn't and shouldn't be based on the actual results. The correct decision doesn't guarantee success. It merely increases the odds that a good result will occur.
While Ross was up I was running through the possibilities of Lester's at-bat. In my mind he was dominating and I didn't want him to come out. I wasn't comfortable at all with the thought of having to get 9 outs from the bullpen, especially when it looked like Lester could get 3-6 more himself.
For me the worst case scenario would have been Ross getting on and loading the bases. No way Farrell could send Lester up with the bases juiced and one out in a tie game. No way. Ross' double was exactly what we needed - the lead, runners on 2nd and 3rd (no routine double-play chance). I agree 100% with letting Lester hit, make his out, and still have Ellsbury come up with a chance to plate one or two.
I get the whole "the results don't make the decision right" argument, but I wanted Lester pitching the 7th.
(After the 7th, I also wanted him pitching the 8th. And I would have even left him in against Adams, though in the postgame Ross said something about Lester hurting himself in the 7th and gutting out the 8th - with that info I would have changed my mind, but I didn't see that watching the game on tv.)
I was listening on the radio at the start of the seventh, when Lester was due to bat fifth, and at that time, the announcers said that no one was up in the Red Sox bullpen. I don't know if or when Farrell had anyone warming at any point during the top of the seventh. Can anyone provide that info?

It seems to me that Lester was going to bat, regardless of the situation, and Ross' double bailed Farrell out. In any case, after the double, in a 2-1 game, what does Farrell have to lose by letting Lester bat? The Sox now had all the momentum. Granted, a base hit in this situation might put the game away, but let's look at other scenarios:

1. Let Lester bat, and he makes an ineffectual out. Red Sox have the lead and one more chance to increase it. This is what actually happened.
2. Let Lester bat, and he somehow produces a run with a gork over the drawn-in infield or something weird happens (not that anything weird has happened in this WS). If Lester somehow drives in a run, that is a crushing blow to the Cardinals and the momentum is all Red Sox.
3. Pinch-hit for Lester with Mike Napoli or Mike Carp. Batter drives in a run or two. More Sox momentum, but they have to piece together five outs from the iffy part of the pen.
4. Pinch-hit for Lester with Napoli or Carp, and the PH does NOT get the run home. This is a potential momentum swing back to the Cardinals, who no longer have to face the LHP that has been shutting them down. Furthermore, maybe if the Sox PH, the Cards bring in Martinez, and frankly, the Sox would rather face Wainwright here.

From the viewpoint of managing expectations and keeping the momentum, the Sox expected Lester to make an out. Anything that happens after he makes the out is gravy. But if Napoli fans, it's a momentum swing for the Cards.

Everything changes if the bases were loaded. I predict that if the bases were loaded, Lester would have been instructed not to swing under any circumstances, due to the risk of a DP. I believe Farrell had already decided to give up that out regardless of the circumstances.
Lichtman's got pretty math to say it was a bad decision, but all he's got to work with is large-sample widespread data. What he does NOT have, is in-the-moment, specific data. Farrell got to look Lester in the eye. Presumably also the Sawx had pitch velocity and location data that they were tracking during the game that gave them a detailed picture of how Lester was really doing (and better numbers from the past games concerning their relievers), as opposed to general "how most pitchers fare under these condtions" numbers.

I think it's safe to say that the 3% difference was well within the error bars of the data in play.

You want to say he should have made the other move, fine. You want to back it up with math, fine. But don't try to tell me that this is some sort of "worst decision..."
"we know that Farrell is understandably wary of going to anyone in his bullpen but Koji Uehara"

Explain to me again why the Red Sox are carrying 11 pitchers.
This has been some of the worst strike zones in a World Series I've seen in years.
There would be no relievers left for either team if both managers followed all the advice here to yank the starters at first wiff.

Getting the lead without having to take Lester out was the desired objective all along. As soon as the Red Sox got the lead there was no way Lester was coming out. That was a case closed.

That walk to Drew almost assured Lester would bat. A run produced from that spot with a PH would at least have gotten Lester out of the game.

That is why the Ross double was so devastating. It got the run in without having to PH for Lester.

I saw no criticism for letting Lackey pitch the 8th the other night. That seems a lot more egregious that letting Lester stay in with the lead.