October 21, 2014
World Series Preview: Giants vs. Royals
On July 28th, the Giants were four games behind the Dodgers, the Royals were five behind the Tigers, and PECOTA put their combined odds of winning the World Series at 4 percent. Neither was a preseason favorite to win the division, neither won the division, neither won 90 games, neither has an MVP candidate or a Cy Young candidate. Neither team's manager will win manager of the year, and neither will be the favorite to win a division going into next year's season. They are a combined 16-2 against the postseason gantlet, and PECOTA puts their combined odds of winning the World Series at 100 percent.
Lineups (AVG/OBP/SLG/TAv, WARP)
The raw numbers for the Giants lineup won't wow anybody, but a lot of the flash is shaded by AT&T Park's stifling effects. As those True Averages show, every starter was better than the league average except Panik, who was roughly a hit-by-pitch and a hustle double away from being included. They had the third-most road home runs in the National League and the third-best road OPS. For all the skepticism inspired by their second-half wilt, the Giants' offense didn't get worse in the summer—it just got different, shifting from a homer-driven offense that put up a sluggish .243/.303/.385 line to a more—well, a more Royals offense, I guess, hitting .271/.321/.393 with little power but 8 percent fewer strikeouts. It's not an extreme offense in any way, with groundball, fly-ball, line drive rates all around average. They had one of the league's best BABIPs, but that again is more about their spacious ballpark than any inherent skill—on the road, they found the average again.
Pablo Sandoval is essentially a left-handed hitter—his career split amounts to more than 40 points of True Average, and in 2014 particularly he found himself wandering dazedly through Ryan Howard's fever dream—which makes this a very lefty-tilted lineup. The results of that tilt weren't pronounced during the regular season—but, then, the lineup wasn't all that tilted during the regular season, with a lot more Morse and Pagan, and a lot less Panik, Belt and Ishikawa. From an analytical perspective, the Giants' postseason struggles against lefties (.204/.306/.222) aren't much more than a small sample. From a narrative perspective, it's a half inning of on-screen graphics and Harold Reynolds worrying on the day(s) that Jason Vargas or Tim Collins pitch.
The Giants will finally get Morse into the lineup as a starter during the AL-park portion of the series. As aggravating as it can be sometimes to see the Royals' best hitter in the sixth spot, their lineup is otherwise a lot closer to optimized than that of the Giants, but with Morse joining the group, Bochy could opt to shuffle his lineup around: Pence has batted first and second for extended stretches this year; Morse typically batted fifth. The problem with moving Pence up (say, to second) and Panik down (to, say, the ninth spot): It would create a run of five left-handed batters in a row, and the Giants would be without a real right-handed threat off the bench. That would be more problematic in a series with Randy Choate, and given the rigidity of the Royals' bullpen—Bochy can say with pretty much certainty that it ain't going to be Finnegan coming into the eighth to play matchups—maybe it's not worth worrying about.
You could call the GIants station to station, or you could just say they don't give away outs: No NL team made fewer outs on the bases (excluding pickoffs or caught stealings); no team was picked off fewer times; no team laid down fewer sacrifice bunts; and only three teams were caught stealing less often.
The Royals got this far into October by doing what they do well, and doing what they don't do well, well. The former: The Royals had the best baserunning team in the majors, by a factor of two; they put the ball in play, with the lowest strikeout rate; and they combined those two qualities in a way that made each play up, avoiding double plays, advancing extra bases, etc. They've been very aggressive on the basepaths this postseason:
Even accounting for three caught stealings, the stolen bases added about a fifth of a win to the Royals' WPA. (The threat of the stolen base might arguably have added still more.)
Another one of those "more than every other team combined" fun facts: The Royals have more sacrifice flies in this postseason than—well, than every other team combined.
Along with all that Royals offense, though, has been the addition of the longball, and about 20 extra points to their isolated patience, each of which has helped the Royals improve their season slash line despite facing postseason pitching. They've homered every 35 at-bats this October, compared to every 58 at-bats from April through September. Of course, even at every 35 at-bats the Royals have actually homered less often than the average team has this October.
The patience has been interesting. Pitches per plate appearance:
Of course, Pablo Sandoval is seeing more than four pitches per plate appearance this postseason, leading the Giants. Stuff happens.
The Royals have used fewer pinch-hitters this postseason (two) than the A's (three). This isn't anything unusual for Yost. In the BP Annual last winter, Chris Jaffe noted that Yost is the perennial favorite for complete games by his position players, and this year his Royals finished last in the league in pinch-hitting appearances by about a two-month margin. There's not a left-hander on the bench, besides Dyson, who—well, let's just say there's not a left-handed hitter on the bench. Of course, things get different in NL parks, and Butler (and perhaps Willingham) will get a shot to bat for pitchers late in games. It'll be interesting to see whether Bochy saves a lefty for Moustakas, who is extremely vulnerable to that lefty funk, but who is also the one power lefty in this lineup Yost could pinch-hit for, setting up a Butler/Lopez or Butler/Affeldt matchup. Even this year—even this year!—Butler hit .321/.387/.460, and his platoon splits have always been country breakfast-sized.
Of course, Yost will use his bench—just for different means. Depending on when Aoki's spot comes up in the order, and whether Aoki reaches, Jarrod Dyson will generally enter in the sixth or seventh inning as a pinch-runner or defensive replacement when the Royals lead. This turns the Royals' defense into something they'll write books about someday, books with unhinged asides about cryptology and the gold standard and the assassination of Marien Ngouabi. Yost can get away with burning his best pinch-runner like this because Dyson isn't his best pinch-runner: Terrance Gore, the fastest man in the AL, has now appeared in 14 career games as a pinch-runner and stolen eight bases without being caught. Interestingly, he didn't steal in either chance he had during the ALCS, each time with Zach Britton—a lefty—on the mound. But Gore did steal thrice against lefties in the regular season, so no reason to think that's an issue.
Colon and Willingham (and, probably, Kratz) exist primarily to replace whoever Gore runs for.
Juan Perez is a defensive specialist and one of the very worst hitters in baseball. Bochy started him against left-hander Gio Gonzalez, the only time the Giants have faced a lefty so far. Would he do it again against Vargas in Game Four? He might instead devote his energies to prayer, specifically for Mike Morse's tender oblique. Matt Duffy got down a clutch sacrifice bunt, but hits like Juan Perez. Other than Morse, Susac is the best hitter on the bench, but Bochy has been cautious about bringing him into games, as most managers are about their catching safety net. Unless Morse is able to start in left, the Giants won't carry any lefty on the bench, setting up an epic Joaquin Arias pinch-hitting appearance against Wade Davis to look forward to.
It's also possible the Giants could carry Gary Brown as the fast guy, likely instead of Duffy. Brown was on the wild card and NLDS rosters but he hasn't been brought in to run yet, showing either a gap between Bochy's strategy for his bench and Yost's, or a gap between Brown's speed and Gore's. (It's a little of both.) Brown appeared once in the postseason, in the 18th inning. If he makes the roster and a game goes 18 or more innings, you'll see him.
Bumgarner has already thrown 32 innings in this postseason, and his next inning will push him over 250 for the season, but there's no sign yet that he's wearing down: His average fastball velocity is actually moving up, from 92.5 mph in July to 92.9 mph in August to 93.1 mph in September to 93.8 mph in October. His only two starts of the year with an average heater 94 mph or hotter: the Wild Card game, and the clinching Game Five of the NLCS. As we noted before the postseason began, Bumgarner's four-seam fastball is one of the few "80" pitches in the league—a whiff rate that is three standard deviations higher than average. It's arguably getting better as he goes deeper into the season: The whiff rate has shrunk some, but he's throwing it for more strikes, and nobody is squaring it up. Opponents this October are hitting .132 against his four-seamer, with a .132 slugging percentage, which means—well, you know what that means.
After five postseason starts, Ryan Vogelsong's magic wandoo finally ran out in the NLCS. Like Bumgarner, he has added velocity late in the season, topping 95 mph against Washington—he had thrown only two pitches harder than 94 mph since the start of the 2013 season. Bochy will stick with him in Game Four instead of replacing him with Yusmeiro Petit, who has thrown nine dominant innings in long relief, and who struck out 46 (against just five walks) in 37 innings over six late-season starts.
Hudson has regained the exceptional control that carried him through a string of dominant April and May starts; it's been two months since he walked more than two in a game, with all but one of those starts including one walk or fewer. Peavy has now made seven postseason starts and is still trying to get the 18th out in one. Of course, if postseason history in a small number of starts was all that important...
...the Royals would be considering replacing James Shields in the rotation. They're not. Shields will start Game One and you'd imagine that there might be talk of bringing him back on short rest if the Royals find themselves down going into Game Four or Game Seven. He has leaned heavily on the cutter during this postseason, and opponents have been elevating it more than usual. That doesn't sound so bad—that outfield!—but between the three home runs and the .360 BABIP (behind 36 percent line drive rate) he's been hit fairly hard.
Like Shields, the Royals pitchers have been carefully protected from third or fourth trips through the order. Only once (Ventura, ALDS Game Two) has a Royal worked even into the seventh, and only once (same) has a Royal thrown more than 95 pitches. Royals have been pulled after five innings/one run, five and a third innings/one run, and six innings/two runs (twice). As a result, is the group fresher than the Giants, who have pitched into the seventh six times, and into the eighth four times? Does this make them stronger than the Giants? Sure. Sure, let's go ahead and say they are. Keys to the Game are presented by Ford. Step into a sleek, stylish Ford Fusion today.
Relief Pitchers (ERA, Innings, FIP)
There's a great nugget in Andy McCullough's BP chat from Friday:
William (Spokane): Is Ned Yost just inconsistent or is he getting better?
What does optimized bullpen deployment mean? Well, in this case, it means this: During the ALCS, the Royals used non-Herrera/Davis/Holland pitchers for just 1 1/3 innings, total. Partly, that was possible because the pitchers were working into the sixth inning reliably, and because the Royals were always ahead. But it's also because Yost went to Herrera in the sixth inning twice, because he used Wade Davis for a two-inning stint. If Yost goes to Holland in a tie game on the road, they might just go rip Casey Stengel's plaque out of Cooperstown and put Yost's right in that spot.
Notable: Holland still hasn't pitched more than an inning this postseason, or this season, or anytime since August 2012. It's a staple of World Series champs that, at some point, the closer comes out of the bullpen in the eighth, but the decision will be a bit more complicated this year based on Holland's lack of track record.
Bochy's bullpen usage will be as unrigid as you'll see. This postseason he has used LOOGYs in the fifth and sixth innings, he has pulled his closer mid-inning (before said closer had allowed a run), he has used his closer in a tie game on the road, he has used a LOOGY for multiple innings at a time, and he has passed over his established leverage guys to put ballgames in the hands of a rookie with seven career innings. He has also skipped his bullpen altogether, letting his starters work deep into games.
Here's the craziest thing: The Giants bullpen this October has allowed seven runs, and seven home runs. Craziest! Even with those homers (four of them against Hunter Strickland), they've got a 1.78 ERA against, though their FIP looks a lot worse, right around 5. They probably won't allow seven home runs every 30 innings going forward. The ones they do will probably have some runners. And the ones they don't will probably have some runners on, too. Though I guess "never underestimate Bruce Bochy and a bullpen in October" isn't a bad policy.
The Giants' defense will focus on keeping the Royals baserunners stationary. Posey has a strong arm and most of his starters will give him a chance—Bumgarner in particular will vary his hold times or shave .3 seconds off his time home unexpectedly. Here's how baserunners have done against each starter over the past two years:
Posey is a very good framer, if a tier below the Lucroys of the world. He and Bumgarner were the 12th-best battery in baseball this year, as far as framing goes, with five runs prevented. Perez and Ventura were among the 10 worst batteries by the same measure.
The Giants don't shift as much as other teams, but they do it more effectively than almost any other team. The Royals shift slightly more often but to little effect. Each team's ranks on some advanced defensive leaderboards:
DRS: Giants 18th, Royals 4th
It is, however, worth pointing out that the Royals are built to be a team that is very good when it is already ahead. It plays good defense that gets better late in games with the lead; it has three excellent relievers who come in with a lead. What we've seen has been what the Royals look like when they lead, and it's great. What do they look like when they trail? Well, there's nobody good on the bench, there's not really a home run threat in the lineup, stealing bases down by three ain't the life for me, and the bottom of the bullpen is just okay. I don't know if this point goes in the managerial section, but it has to go somewhere.
Bochy's managerial tendencies have been touched on throughout this preview. He's a great regular season manager—Chris Jaffe has found that his players, especially veterans, have simply done better than expected under his guidance—and he's working on an eight-series postseason winning streak (plus a Wild Card win). It'll take a longer article than this to get to the question of why, or how, but one theme of the three Octobers has been that Bochy gets valuable performances out of guys we've forgotten about. In 2010, Edgar Renteria went from a bench player at the start of the postseason to a hero by the end. In 2012, Tim Lincecum lost his spot in the rotation and ended up saving his Drinks Free For Life In This City privileges by pitching so effectively out of the bullpen. In 2010, Zito was left off the postseason roster entirely; in 2012, scarcely improved as a pitcher, he was back in the rotation—then lost his spot in the rotation after a terrible outing—then found his way back to the rotation, where he was a hero. Already this postseason, Bochy has wrung value out of a bench that is, on paper, irredeemable. Is that Bochy? Who really knows.
Thanks to Kate Morrison for assistance.