So You’re Starting Jeremy Guthrie In The ALCS

When the Royals and Orioles resume their rain-interrupted ALCS tonight in Kansas City, the Royals are going to use their fifth-best starting pitcher, and on the surface this doesn’t make a lot of sense. Not that Yordano Ventura was going to pitch on two days rest, or that James Shields was likely to go on three days rest, but with Jason Vargas and Danny Duffy around, Ned Yost has options. Instead, he’s going with Jeremy Guthrie, who hasn’t pitched since Sept. 26, a full 17 days of rest. Though he’s 35 years old, Guthrie has never thrown a pitch in a postseason game, so the often-seen “but playoff-tested” excuse doesn’t work here.

When we say “fifth starter,” that’s based entirely on performance. Of the five regular Kansas City starters, Guthrie’s ERA this year was the worst. His FIP was the worst. His WAR, despite throwing the second-most innings, was the worst. Over the last five years, his FIP is 215th of 228 qualified pitchers, basically making him Randy Wolf with fewer injuries. Just two seasons ago, he was getting traded straight-up for Jonathan Sanchez after a brutal half-season in Colorado. Now, he’s starting a playoff game, one of the most important games for his franchise in years.

After years as a quietly acceptable innings eater for Baltimore and Kansas City — regularly pitching 180-200 innings despite a homer problem and mediocre strikeout numbers, and never topping 2.5 WAR — Guthrie will now have the eyes of baseball upon him as the O’s try desperately to beat their old friend and climb back into the series. A 2-0 series lead gives you some leeway, but this also isn’t exactly an ideal scenario for Royals fans, either.

Of course, while Guthrie is easily the fifth-best Royals starter, this isn’t a choice between five pitchers. Shields and Ventura are unavailable, obviously, and Duffy, who insists he’s healthy, has been given the Shelby Miller Memorial “Milk Carton In The Playoffs” award, so this is really a choice between Vargas, who started Game 1 of the ALDS, and Guthrie. One’s generally agreed to be a better pitcher. The other is starting tonight. So how do we parse this?

I thought about looking into the worst starting pitchers to make a postseason start, but that’s both unfair to Guthrie and unsatisfying to us, because quantifying such a thing is so iffy. Hong-Chih Kuo, for example, started Game 2 of the 2006 NLDS for the Dodgers despite having only five uninspiring major league starts under his belt. Though he never made it as a starter, Kuo did go on to have a few dominant years as a reliever before finally blowing out his arm for the umpteenth time. So instead, let’s go about this the other way, and try to pretend we’re Yost for a minute. Here’s why this might work.

First, there’s this: Yeah, Vargas is probably a better pitcher than Guthrie is, but the difference just might be so, so minimal. The chart below shows their season stats — multiple seasons would be better, probably, but let’s just do it this way to reduce differences in park effects, as this is Vargas’ only year with the Royals — and there’s more that’s similar that isn’t.

Name	HR/9	K%	BB%	K-BB%	AVG	WHIP	BABIP	GB%	LOB%	ERA	FIP	E-F
Jason Vargas	0.91	16.2%	5.2%	11.0%	.265	1.27	.299	38.3	74.5%	3.71	3.84	-0.13
Jeremy Guthrie	1.02	14.4%	5.7%	8.7%	.268	1.30	.294	43.6	72.4%	4.13	4.32	-0.19

Both out-pitched their FIP by a small amount, thanks to a nearly-identical BABIP that’s within career norms. As far as walks and left on base percentage goes, both have been about the same. For the minimal usage that batting average against provides, that’s essentially identical as well. Vargas’ main advantages are that he gets some strikeouts where Guthrie doesn’t, and he keeps the ball in the park a little better. I would usually say that getting more grounders is a point in Guthrie’s favor, but after watching the Lorenzo Cain road show, I’m not so sure that fly balls in front of that Kansas City outfield — merely noting Cain sells Alex Gordon and Jarrod Dyson short — are such a bad thing.

Vargas is tentatively scheduled to pitch Game 4, so the argument could have originally been made that this doesn’t really matter, because both would start anyway. (I’d still like to make the case for Duffy, who was outstanding this year, but that one-pitch shoulder-related outing in New York in September and reports that the team is concerned about his mechanics are enough to satisfy those concerns.) But with Monday’s rainout pushing things back, that’s no longer so clear. Shields would be on regular rest for Game 4, and Yost indicated he was considering replacing Vargas. That could mean Vargas never pitches at all, because Shields could come back on short rest for Game 7, or the series could be over.

Over the course of a season, you probably want Vargas, because more strikeouts and fewer homers will add up over six months. Over the course of a single game, it probably doesn’t matter so much. (Even Steamer’s 2015 projections don’t see an enormous difference, pegging Guthrie to a 4.60 FIP and Vargas to a 4.35. Neither of those are great.) So instead, you look at what does make them different, and you go with the obvious: Vargas is a lefty. Guthrie is a righty.

That’s a not-insignificant fact, either, because there’s some serious platoon splits here:

wOBA	2014			Last three years
	v LH	v RH		v LH	v RH
Guthrie	.364	.270		.377	.294
Vargas	.296	.321		.314	.318

For some context there, Jayson Werth had a .377 wOBA this year. Buster Posey had a .371. That’s basically what Guthrie has allowed lefties to do to him over the last three seasons. Meanwhile, righties hit him this year like Allen Craig hit everyone, and Allen Craig hit no one. While Vargas has shown a much smaller split overall, Guthrie is particularly well-suited to face a right-handed lineup. Against righties this year, Guthrie had the same wOBA against as Chris Sale and Jordan Zimmermann. Against lefties, he was Nick Martinez or Kevin Correia. It’s not a one-year sample size thing, either, because that’s been his issue for years.

That, right there, is why this could work, because with Chris Davis out, the Orioles don’t really have a dangerous lefty bat. Against the righties Shields and Ventura in Games 1 & 2, six of the nine Orioles were righties, excepting only Nick Markakis, Alejandro De Aza, and Ryan Flaherty, all, at best, league-average bats, and Flaherty considerably below. Meanwhile, Steve Pearce has a massive platoon split, both this year and his career, that has him most productive against lefties. So does Nelson Cruz. Adam Jones does this year too, though that’s rarely been true in the past. The problem for Guthrie is that being “better against lefties” isn’t the same thing as “being bad against righties,” though it’s in his favor more than it would be against Vargas.

We can’t know what the layoff will do to Guthrie, though we’ve seen that be an issue to others in the past. (Ricky Nolasco, last year, made his final start on Sept. 25 and then ran into trouble on Oct. 15, a similar length as Guthrie will have.) Guthrie has made 31 starts on more than six days rest in his career, and his performance has been all but identical to his usual four days rest.

So it’s Guthrie, and that makes some sense. He’s much better against righties than his overall line would indicate, and Baltimore’s a righty-heavy offense. He’s trying against lefties, though. While it hasn’t really worked out overall — I mean, just look at the wOBA marks above — there’s at least something interesting about him against lefties.

Via Brooks, here’s his raw whiff counts from the last four years on his change, against lefties:

2011: 30
2012: 25
2013: 23
2014: 78

Again, to reiterate the level we’re talking about, Felix Hernandez had 25 on his change against lefties in June alone this year, but let’s give some credit, because the 78 Guthrie had this year equal the last three years combined. Those are raw numbers, not percentages, and that’s despite the fact that he really didn’t use his change much more against lefties than he did last season. (He did, however, limit his sinker in favor of more four-seamers.)

If Mel Antonen is to be believed, Shields’ value to the Royals goes beyond what he’s done on the mound:

Shields taught Guthrie a new grip on his changeup and that is an improved pitch, Guthrie says.

Here’s how that change looked to a lefty, in this case against Kennys Vargas in August:

It’s not a bad little pitch, as far as pitches go. It doesn’t do a ton, but it moves a little. He’s getting some swings-and-misses against lefties with it, more than he had before, and if Shields is in some way to thank for that, all the better for Kansas City. It’s at least a weapon he can use, something to counteract the fact that lefties crush his fastball, and they crush his slider. It’s something, anyway.

That Guthrie probably is the best choice to start this this game may say more about the general existence of Vargas and Duffy’s late season issues than anything else. Fifth-best starters shouldn’t generally be pitching Game 3 of the ALCS, but that’s what this situation has turned itself into. As long as Guthrie can avoid big mistakes — not only in terms of homers, but in the sense that he made more throwing errors this year than any pitcher since 2002 — he might just get through this. At this point, nothing that happens in a Royals game would surprise me.