Optimizing the D-backs Lineup: Bat Goldschmidt Second or Fourth

In sabermetric circles, you frequently hear that the ideal lineup is worth only five to fifteen runs over the average lineup. But five to fifteen runs is more than nothing, even over the course of the season, and lineup construction also happens to be fun to talk about. My preferred lineups — and justification — are below.

Lots to talk about here, but let’s first get on the same page in terms of who should actually be in both of these lineups.

Who’s playing?

I covered this yesterday, but the D-backs have an unbelievably wonderful opportunity to match Cody Ross and Eric Chavez in a platoon while Mark Trumbo‘s foot heals. It’s rare to have any “half time” players, let alone two. And here, the two players can trade off even though they don’t play the same position, thanks to Martin Prado. They hit from different sides of the plate, and both players happen to have significant platoon splits. There’s really no reason not to do this, as Ross tries to return from injury and Chavez tries to prevent one.

But they really do have strong platoon splits, and so long as they’re deployed in such a way as to take advantage of that, they’re both very strong hitters. Ross’s triple slash against LHP last season was .391/.430/.582, which is even more pronounced than his career splits versus LHP (.296/.360/.571). And Chavez has already been consigned to this status, although he happens to be damned good versus RHP (.275/.354/.500 for his career). As platoon players, Ross and Chavez might be better than any hitter on the team not named Goldschmidt (although Hill might give Ross a run for his money against LHP). We don’t know if or when Ross will return to his previous form, but I’m assuming he will — and the only way to find out if he can fully come back is to play him.

The only playing-time choice I expect might be controversial is actually playing Gerardo Parra versus LHP. I know, Parra is not very good against lefties, especially as compared to his RHP numbers (if he only faced RHP, Parra would probably be a superstar). Last year, Parra wasn’t a complete embarrassment in terms of getting on base (.276 OBP vs. LHP), but his slugging percentage was putrid (.226 SLG). Keep in mind three things, though: 1) the next-best option is Tony Campana; 2) Parra is trying out a new stance which may help him against LHP, and giving him PAs is the only way to find out; and 3) Parra provides brilliant defense regardless of the handedness of the opposing pitcher, making him more valuable than most hitters with numbers that bad.

Basic lineup optimization

I’m not going to try to recreate all of the excellent work on lineup construction, but ex-Beyond the Box Score great Sky Kalkman’s summary on how to apply the lineup lessons of The Book is well worth three minutes to read. For our purposes, let’s adopt Kalkman’s grand finale:

Here’s how the lineup spots rank in the importance of avoiding outs:

#1, #4, #2, #5, #3, #6, #7, #8, #9

So, you want your best three hitters to hit in the #1, #4, and #2 spots. Distribute them so OBP is higher in the order and SLG is lower. Then place your fourth and fifth best hitters, with the #5 spot usually seeing the better hitter, unless he’s a high-homerun guy. Then place your four remaining hitters in decreasing order of overall hitting ability, with basestealers ahead of singles hitters.

It’s that simple, folks. The hardest piece of this to swallow may be the lack of emphasis on the three-hole, despite it normally being thought of as the most important of lineup slots. The explanation for why it’s not is that it’s the lineup spot most likely to come up with two outs and no one on — and that discount is great enough to make up for the fact that the three-hole will come up about fifteen times more per season than the cleanup spot.

Before getting to the D-backs’ actual players, there’s one other thing to touch on — putting the pitcher in the 8th spot instead of the 9th spot. The Book teaches that having the better hitter hit closer to the top of the lineup is worth more than having the 8th spot come up more frequently. But this is also the National League we’re talking about, and the pitcher’s spot won’t be filled by a pitcher for all of the plate appearances in a particular game. Beyond that, there’s the fact that both of the guys I’ve put in the 9th-hole have severe platoon splits, and there may be a benefit to having them bat later — they’re more likely to hit against an opposite-armed pitcher. Say a lefty is starting, but running out of gas when the 8th spot comes up. The manager puts in a pinch hitter, opting for a right-handed batter. That might increase the opposing manager’s willingness to bring in a RHP to face the pinch hitter, and if that phenomenon happens at all, it helps justify putting a professional hitter 9th, after the pitcher.

Different spots against different hitters?

Here’s my lineup table again, but this time with on base percentage (avoiding outs) and slugging percentage, the two factors that should matter most in this type of analysis. For most of these players, I’m using 2013 statistics. For Miguel Montero, I’m using career splits because while last year counts, I think it’s wildly different than his true skills. And for Chris Owings, I’m using his ZiPS projection — except that I’m walking both numbers down 20 points against RHP and up 30 points against LHP, as a simulated platoon split. Overall, by the way, six of the eight D-backs hitters do quite a bit better against lefties. So don’t go crazy when you see Prado’s .473 SLG “wasted” in the leadoff spot — the slugging at most spots goes up.

Overall, the D-backs’ two best hitters are, without question, Aaron Hill and Paul Goldschmidt. But a significant if intuitive thing happens when they face LHP — their slugging percentages go through the roof. Goldy’s OBP was actually better against RHP last season than against LHP, but the OBPs for both players didn’t have a significant platoon split.

The difference in slugging, however, made them different hitters against LHP. If the most important three hitters should be in the 1, 2 and 4 spots, OBP should get front-loaded, SLG behind. Goldy is probably most valuable for his OBP against right-handers, but even then, his slugging percentage is the best on the team. Sounds like a two-hitter under a saber regime. Against lefties, his slugging is ridiculous, and it takes center stage, pushing him to the cleanup spot (it helps that Ross rakes against LHP also).

But… what?? Parra and Pollock in the three-hole?? Yes — but remember, we’re saying the three-hole is the fifth-most important spot in the lineup. This is the spot that would probably feature Mark Trumbo in a saber lineup, because home runs form the totality of his game at least as much as for any other hitter in the majors, and that’s where you put that type of guy. But Parra and Pollock are both very good hitters against opposite-handed starters, good enough to be the fifth-best hitters on their team in those circumstances. Hitting them this high has a secondary benefit of increasing the number of plate appearances they’re likely to get against the starter — against whom they keep a platoon advantage.

As for the other spots, Chavez and Montero are kind of a wash against RHP, but I like Chavez cleanup because if he gets lifted later in the game (Ross pinch hit?), the 2-6 spots would then go L-R-L-R-L-R. Prado and Hill, our two leadoff guys, get bounced pretty far down the order against same-handed SPs. But Prado might have the third-best OBP on the team against LHP, and the other two guys (Goldy and Ross) have significantly better slugging percentages. Same can be said for Hill re: RHPs.

What do you think? What would your two lineups be?