Sunday, November 2, 2014

The Best and Worst Players of the 2014 Postseason

The 2014 postseason was a strange one. It started off with one of the greatest single games in playoff history, and contained a number of other exceptional contests, including one that tied the record for most innings in a playoff game – but the series in which those games occurred were predominantly brief. Through the LCS round, none of the six multi-game matchups included more than one win by the team that lost the series. The World Series broke that trend, lasting the full seven, but it did so while having the average game become far duller than those involved in the briefer clashes that preceded it.

On the other hand, the individual performances seen in October 2014 were often quite memorable. And this space has previously explored Championship Probability Added (CPA), which is designed to analyze just those sorts of performances. So… let’s do that!

In case you aren't aware of this already (that is, in case you didn’t watch the playoffs at all this year), you should probably know up front that this October’s most interesting performances came from pitchers. So we’ll start by covering the comparatively dull hitters, then finish things off with the moundsmen.

Bottom 5 hitters:

Joe Panik
Brandon Crawford
Buster Posey
Nori Aoki
Salvador Perez

All of the players selected were World Series participants. This is pretty normal (since World Series games are more important, and thus weighted more heavily by CPA), especially since the Fall Classic significantly outlasted every other playoff series (and thus had the even-more-crucial games 5, 6, and 7 to factor in).

The three up-the-middle Giant hitters were all at least relatively neutral performers through the first three postseason rounds; they then put up scores of -.116, -.091, and -.094, respectively, in their last appearance. Posey’s struggles were particularly noteworthy, both because of his status as one of the team’s best players and because he’d played at least something of a positive role in each of their prior two titles.

Nori Aoki, like his San Francisco counterparts, was unspectacular but not damaging through the LCS. Then came the World Series, in which he went 1 for 14 and hit into a pair of double plays. His Game 7 effort was particularly harmful; he went 0 for 3, and two of the three outs came with the tying or go-ahead run in scoring position.

But none of them hold a candle to Salvador Perez. Perez’s raw numbers are rather dreadful (.207/.233/.276), but they also don’t tell the full story; his World Series slash line was actually rather good (.333/.360/.500), and yet that’s also where the worst of his damage was done, -.237 of his total CPA of -.308.

His Series performance wasn’t all bad. In Game 1, he became the first (and thus far, only) hitter to produce a World Series run against Madison Bumgarner with a seventh-inning homer. Of course, that homer made no difference in the outcome of the 7-1 loss. Game 2 went better, as Perez drove in a pair of runs in a win. Game 3 was another Royal win, but an 0-for-3; Game 4, on the other hand, went to the Giants despite Perez’s three hits. Game 5 (1 for 3 in a shutout loss) moved the needle very little, but Game 6 (2 hits and a run in a blowout win) was worth +.091 WPA, and thus +.0455 CPA, pushing Perez slightly into positive territory for the series to that point.

And then came Game 7. Perez’s first showing was a fairly inconsequential HBP from the hand of Tim Hudson near the end of the Giant starter's tenure on the mound. His second chance came in the fourth, with San Francisco just having taken the lead, but KC having put its leadoff man on base. Perez promptly hit into a rally-squelching double play. In the seventh, he led off with a flyout. And in the ninth, with the tying run having miraculously appeared at third and two away, Perez fouled out to end the season.

The DP and the game-ender were both rather harmful. Put them together in the absence of any significant positives, and you get a WPA of -.272, in probably the most important game Perez will ever play.

That’s why he’s at the bottom of the list. It may not be fair, but the MLB postseason has minimal association with fairness.

Top 5 hitters:

Alex Gordon
Pablo Sandoval
Hunter Pence
Mike Morse
Matt Adams

Adams is our first non-World Series participant, and he demonstrates roughly what you have to accomplish to make a list like this when the LDS and LCS rounds are less dramatic than might be ideal. The Cardinal first baseman hit a pair of crucial late-inning homers – the first of them in the seventh inning of NLDS Game 4 off of Clayton Kershaw to put St. Louis in front, and the second (somewhat less surprisingly) against Hunter Strickland in inning eight of the NLCS’s second game, breaking a 3-3 tie. Had it come in a series his team won, that NLCS homer would have been considered much more memorable; as it is, it stands not-quite-alone as a Cardinal highlight in an otherwise bleak series for the Redbirds.

Morse appears on this list despite having missed the postseason’s first two rounds. He returned for the NLCS, and tied the fifth and final game of that series with a pinch home run in the eighth. Morse then sealed his spot on the top hitters' list in the closing game of the next series, driving in a pair of runs in a Game 7 that his team won by one.

Sandoval and Pence are remarkably close here, which makes sense, because if a hitter had won the World Series MVP, it would have been very difficult to choose between them on the standard stats as well. It was Sandoval who set the single-postseason hits record, and Sandoval who had the big day in Game 7 of the Series (3 for 3 with a double and two runs scored, +.143 WPA), but Pence got better as the playoffs went on, and had four multi-hit games in the World Series as well.

And yet, CPA puts them both behind Alex Gordon, who hit .204/.317/.370 for the postseason as a whole, and .185/.214/.296 in the World Series. So how does THAT work?

Mostly, it’s about timing. Gordon went 0 for 5 with a walk in KC’s epic victory in the Wild Card game, but hit very well in both the ALDS (.300/.462/.500, +.030 CPA) and ALCS (.250/.438/.583, +.071 CPA). Those championship probability figures might not seem like much, but remember that the Royals swept both series, making it hard for their players to rack up truly prodigious probabilistic performances. Gordon was their second-best hitter in the LDS, and their best in the LCS.

Then came the Series. Gordon scuffled for most of it, going 3 for 24 through 6 games, though he did drive in a run and score another in KC’s one-run victory in Game 3. But in Game 7, Gordon more than made up for his early faults. In the second inning, with the Royals trailing 2-0, Gordon hit an RBI double, then came around to tally the tying run on a pair of flyouts. With KC down a run in the fourth, he led off by getting plunked, only to be erased on a double play. And with two outs in the ninth and that one-run margin still holding, Gordon struck a single to center that turned into the most memorable single play of the Series – and very nearly became one of the most famous in baseball history. A miscue or two in the Giant outfield transformed that single into the equivalent of a triple, and spurred what will probably be an unending argument about whether Gordon should have been sent home, or whether he could have scored if he’d run harder, or any number of other aspects of the play that baseball fans will doggedly seek out for bickering purposes.

Gordon was, of course, left at third to end the Series. But his efforts in Game 7 gave the Royals a much better chance than they otherwise would have had, and his star performances in the two earlier rounds put them in position to have that chance in the first place. And that’s what puts him atop the list of this October’s hitters.

Bottom 5 pitchers:

Brandon Finnegan
Jake Peavy
James Shields
Ryan Vogelsong
Tim Hudson

Sixth from the bottom on this list was Hunter Strickland, whose score of -.096 is effectively identical to Finnegan’s. The two pitchers throw with opposite hands, and the lefty Finnegan is four and a half years younger than Strickland – but apart from that, they were pretty much the same guy in these playoffs. Both had brief-but-promising conclusions to the regular season (seven innings each), and both were given mildly important bullpen roles in October based on that. It worked out roughly the same for each of them, and not in a good way. Finnegan pitched 6 innings and allowed 7 runs on 9 hits and 5 walks, which is a 10.50 ERA. Strickland’s ERA was better, though his 7.56 mark isn’t exactly one that his agents will be citing in his arbitration hearings a few years from now. His postseason was mostly notable for the shocking SIX home runs he allowed in 8.1 innings, five of them to left-handed hitters.

The lesson here, if there is one, is that seven good innings in the regular season should not necessarily be used to assign spots on the postseason pitching staff.

The other four members of this group are all starters. Let’s examine James Shields first. “Big Game James” is a man in desperate need of a different nickname; after his rather dismal efforts this October (a 1-2 record in 5 starts, with Kansas City's hitters having bailed him out of two more likely losses), his career postseason ERA stands at 5.46. He remains a fine pitcher in the regular season, but his production in Big Games has, to date, left something to be desired.

And pitchers 4, 2, and 1 on the list? They are three of the four members of the postseason rotation of the world champion San Francisco Giants. Peavy, #4, pitched very well in his first start, Game 1 of the NLDS. Things got much, much worse from there; he was lifted after four innings in Game 2 of the NLCS, then gave up four runs in the second game of the World Series, and then got knocked out after 1.1 in a Game 6 blowout loss. Peavy’s ERA for the 2014 postseason was 6.19; sadly, that establishes a career-best among his four visits to October.

Vogelsong, #2, also did well in his NLDS appearance, helping his team secure the fourth and final game of that series. He made two further starts, both times giving up 4 runs in 3 or fewer innings, though the Giants won both games despite him. After that, he was relegated to a lone inning out of the bullpen; he kept the Royals off the scoreboard in the eighth inning of Game 6, but one doesn’t get much credit for maintaining a 10-run deficit. Vogelsong had been a hero of San Francisco’s 2012 title run, posting a 1.09 ERA that year; two seasons later, his mark was a much uglier 6.57.

Hudson exceeded both of his teammates' negative efforts, though his raw numbers were better than either. Like his two compatriots, Hudson’s NLDS start went well, featuring 7.1 innings of one-run baseball; he proceeded to be overshadowed by the other 10.2 innings that were played in the game. His lone NLCS outing was poor, but not disastrous, and the Giants won anyway. The same could not be said of Game 3 of the World Series, in which Hudson was slightly outpitched by Jeremy Guthrie and absorbed the loss despite a respectable 5.2-inning, 3-run effort.

The real damage to Hudson’s record came in Game 7, in which he lasted only 1.2 innings and blew a 2-run lead. The -.164 WPA was not catastrophic, but it was quite damaging in the most important game of Hudson's distinguished career, and it had to be made up for by someone else.

Fortunately for the Giants, they had just the right person on hand.

Top 5 pitchers:

Madison Bumgarner
Jeremy Affeldt
Wade Davis
Kelvin Herrera
Greg Holland

Yeah, that’s about right.

You may have heard that the Royals had a pretty good bullpen this year. That bullpen showed up for October, and did its level best to carry an otherwise unspectacular KC roster to the world title. The numbers speak very effectively for themselves:

Holland: 11 games, 11 innings, 4 hits, 6 walks, 15 K’s, 1 run allowed. 7 saves
Herrera: 11 games, 15 innings, 11 hits, 7 walks, 16 K’s, 3 runs allowed
Davis: 12 games, 14.1 innings, 8 hits, 2 walks, 20 K’s, 2 runs allowed (one earned)

That’s a combined 40.1 innings pitched, and 5 earned runs allowed – a 1.12 ERA in over four full games’ worth of highly important, closely contested baseball. And even the runs they allowed were predominantly irrelevant. Davis’s two runs came with the Royals leading by six and trailing by four, respectively, and Holland’s lone runner crossed the plate during a save that turned a three-run lead into a two-run win. It’s extraordinary work from an extraordinary trio of relievers.

And 35-year-old journeyman Jeremy Affeldt did better than any of them. The lefty pitched in 11 October games, throwing a total of 11.2 innings, and allowed not a single run to score – including the stranding of all five of the runners who were on base when he entered. The weightiest outing of the bunch, of course, was in Game 7 of the Series, in which Affeldt entered a tie game in the second inning with two runners on, stranded both of them, and threw two more scoreless innings before departing, good for a +.164 WPA in the most important game you can have.

The best thing about Affeldt’s work? The fact that those 11.2 scoreless innings included a counterculturally miniscule total of two strikeouts.

All right, enough about performances that were merely good enough to lead the way in most Octobers. Let’s move on to Madison Bumgarner, who was at least one level beyond that.

There are any number of ways to describe Bumgarner’s unbelievable postseason performance this year. Bumgarner set the record for most innings pitched in a single postseason, with 52.2. In those record-breaking innings, he had a 1.03 ERA. He made six starts, all of which lasted at least 7 innings, and gave up 3 runs or fewer in each of them. He threw two shutouts. He struck out 45 batters, and walked six.

His starting work was obviously great, and would have put him atop this list on its own – but his biggest game came out of the bullpen, as he threw the last five innings of Game 7 with a one-run lead, and kept it intact. It was a stunning effort on two days’ rest, worth a WPA of +.603 that would be exceptional in any game, let alone the last of the year.

If that’s not enough (and it is), Bumgarner did all of this in an extremely hostile environment. Ignoring Bumgarner himself, the next 7 pitchers on the CPA list for this season were all relievers. The next-best starter of the 2014 playoffs had a comparatively paltry CPA of +.060 – and that was recorded by Jordan Zimmermann, whose team lost his only start, and went on to be knocked out in the NLDS. The four starters after Zimmermann were Zack Greinke, Doug Fister, Hyun-jin Ryu, and Matt Shoemaker, with scores ranging from +.035 to +.015, and all of them were also eliminated after one round. There were only two non-Bumgarner starters whose teams made the LCS and who had positive CPA scores – Yordano Ventura and Miguel Gonzalez, at +.014 and +.001, respectively.

All of which is to say, Bumgarner posted an unbelievable playoff run in the middle of what may well have been the worst overall postseason for starting pitching in baseball history. Given that the Giants entered as a wild card team (with championship odds of roughly .0625 when the playoffs began), it would not be entirely unfair to say that Bumgarner, whose CPA was greater than the margin of .9375 that initially separated San Francisco from the title, was just about solely responsible for their winning the World Series.

It is, in fact, a performance so titanically impressive that its impact cannot be fully understood by simply comparing it to those produced by the other players of this season. Madison Bumgarner’s dominant October has to be placed into a wider historical context.

Which is what we’ll do next time.

No comments:

Post a Comment