Optimizing The Yankee Lineup

In recent comments, Joe Girardi suggested that he is considering tinkering with the Yankees lineup:

“We’ll continue to talk about our lineup and how guys fit in,” Girardi said. “Right now we don’t expect a ton of changes. We’ve talked about some things internally that we might try to do. I don’t really want to get into it, but we’re not really sure exactly how our lineup’s going to break… We might play around a little bit in spring training.”

While he is unlikely to make radical changes, I wondered what it might be like if a real saber junkie was filling out the lineup card for the Yankees. I decided to look at how the Yankee lineup would be constructed under two statistical frameworks, using career statistics to estimate performance:

1) By “The Book,” with help from Sky Kalkman’s excellent summary of the relevant methodology:

Another way to look at things is to order the batting slots by the leveraged value of the out. In plain English (sort of), we want to know how costly making an out is by each lineup position, based on the base-out situations they most often find themselves in, and then weighted by how often each lineup spot comes to the plate. 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. Finally, stop talking like the lineup is a make-or-break decision.

Using career numbers does not give recent performance enough weight, so I attempted to account for that at the margins. Where two players were close, I gave the edge to the player who had been more successful in that split over recent seasons. Depending on how much weight you give such factors, these rankings could change wildly, so I am willing to accept any suggestions regarding how to rank the players.

Against righties, I would say that Alex Rodriguez is clearly their best hitter, and then Mark Teixeira and Curtis Granderson are fairly interchangeable in the next 2 slots, with Jorge Posada and Robinson Cano trailing behind them. Nick Swisher, Derek Jeter, Brett Gardner, and Russell Martin round out the lineup in that order.

Against lefties, Jeter, Rodriguez, and Teixeira are the Yankees’ best hitters. Jorge Posada, Nick Swisher, and Robinson Cano are the next group of 3, and Russell Martin, Brett Gardner, and Curtis Granderson finish this lineup. If Andruw Jones plays, he fits into that second group and Cano slides down a slot.

This is how I would order them, based on the rules that Sky laid out in his post:

vRHP
Rodriguez
Teixeira
Posada
Granderson
Cano
Swisher
Gardner
Jeter
Martin

Rodriguez is clearly the best hitter and fits the OBP requirements of the leadoff position. Teixeira got the 2 spot while Granderson is at cleanup due to Teixeira likely having a better OBP. Cano is then placed at 5, and Posada fits nicely in the 3 spot. Swisher, the best remaining hitter, is 6th, while Gardner precedes Jeter and Martin by dint of being a basestealer.

vLHP (assuming the typical starting 9)

Teixeira
Jeter
Swisher
Rodriguez
Posada
Cano
Martin
Gardner
Granderson

VLHP (if Jones plays)
Teixeira
Jeter
Swisher
Rodriguez
Posada
Jones
Cano
Martin
Gardner/Granderson

Based on recent performance, I deemed Teixeira the best of the top 3 hitters, and then placed A-Rod at cleanup and Jeter at #2. The remainder of these lineups fit easily with the evaluations of talent that I made above.

2) The Lineup Analysis tool at Baseball Musings allows you to input statistics for the 9 players and then estimates the runs that would be scored by each permutation. This is the best Yankee lineup against righties:

Posada
Rodriguez
Jeter
Granderson
Teixeira
Swisher
Cano
Martin
Gardner

Against lefties, without Jones:

Martin
Teixeira
Swisher
Rodriguez
Jeter
Cano
Posada
Granderson
Gardner

Against lefties, with Jones:

Martin
Teixeira
Posada
Rodriguez
Jeter
Cano
Jones
Granderson/Gardner
Swisher

I certainly do not expect Girardi to use either of these lineups, despite the fact that they would optimize the team’s ability to score runs. Remember, the ordering of the lineup impacts just a handful of runs per season, such that it is likely not worth the bruised egos to fiddle with established lineup positions. Rather, I expect that Joe will use the following perfectly reasonable combinations, much like he did last season:

Vs. RHP
Jeter
Granderson
Teixeira
Rodriguez
Cano
Swisher
Posada
Martin
Gardner

Vs. LHP
Jeter
Swisher
Teixeira
Rodriguez
Cano
Posada
Granderson/Jones
Martin
Gardner/Granderson

What would your lineup look like?

19 thoughts on “Optimizing The Yankee Lineup

  1. Thanks for the mention. Here’s the link to the Lineup Analysis Tool. I wonder if Girardi might bat Martin at the top of the order in certain situations?

  2. Good stuff, Moshe. I love the idea of Tex leading off, although the Pirates have a better shot at winning the World Series than that ever happening.

  3. […] This post was mentioned on Twitter by Larry Koestler and Peter, moshetya. moshetya said: New post: Optimizing The Yankee Lineup http://bit.ly/fovlf2 […]

  4. stunna4885

    yankees just inquired on tim hudson. braves said no way. derek lowe is avaiable though. apparently yankees dont like what the braves are asking for lowe in prospects. plus he has 2 years at 30 mil left on his contract. i like lowe but only for b level prospect or two. word is cashman is turning over every rock looking for a starter but has had no luck thus far. if we cant get a starter before spring training ends it wont be for a lack of trying.

    • Wow, that’s some pretty deep intel — where’d you hear the Yanks inquired on Tim Hudson? I’m not seeing that anywhere.

      • stunna4885

        from nybd contributor frank russo who has a pretty reliable tampa source.

        • T.O Chris

          I don’t believe a word on Hudson or Lowe at this point neither one is all that viable at the top of a rotation in the AL let alone the East and for what they would cost (money for Lowe, pieces for Hudson) they aren’t worth the return.

          I don’t Cashman is looking but I’ll stop watching baseball for the season if either of those 2 trades go down, that’s how sure I am that those deals aren’t in the works.

          It really just doesn’t make a lot of sense from either side when you consider the Braves are setup to go for it and shouldn’t be listening to trade a starter unless they fall out of it in the season, and as for us we really don’t need some average 3 or 4 starter if we are going to pay money for Lowe or take on an older pitcher who may not do well in the AL anymore like Timmy. Id honestly rather go into the season with what we have and see where teams are at the break, the last thing we want is to make a trade now for someone like Derek Lowe and then have Josh Johnson or someone become available and have to trade more (top end pieces this time) prospects and be stuck with Lowe for another year and a half in the 5th spot fighting Burnett for most overpaid.

    • T.O Chris

      Buster Olney reported today via his blog that The Yankees did not call about Hudson yesterday and he also added that the Braves have no interest in trading starting pitching anyway.

  5. Oops. The link was in there originally, but I had to rewrite the entire post because I originally wrote it without doing it by platoon split, and I must have forgotten. SOrry about that.

    As for Martin, it depends on how he and Jeter hit. Unless Jeter craters, he is likely to lead off, but I can see a situation where Martin gets as high as #2 against lefties.

    • T.O Chris

      You could see Russell Martin hitting second? Swisher would have to tank pretty hard for that to happen and so would Cano because id rather see Cano move to the 2 hole with Arod 3rd and Teixeira 5th if Jeter is moving out of leadoff (and falling past 2nd) for Gardner.

      • If he ups his OBP without recovering power, I could see them putting him 2nd and leaving the power like Swisher and Jorge backing A-Rod and Cano up. Not likely, but I could see it.

        • T.O Chris

          I’d be pretty shocked if he was able to bounce back that far, I love the Martin signing and I think he is going to do well, maybe well enough to keep next season as a trade chip or something but putting him that high up so soon might not be the best idea. No need to add a lot more ABs to what he is already going to be doing with catching duties.

          I also just really like the Combo of Swisher Vs LHP and Granderson Vs RHP hitting 2nd because I think it gives us the deadlies combo in that spot and both player benefit quite a bit from hitting ahead of Teixeira and Arod, I know some don’t believe that but they both clearly get more fastball and we all know Granderson struggles more with breaking pitches.

  6. Brads

    My line-up would be
    Gardner
    Cano
    Jeter
    ARod
    Tex
    Posada
    Swisher
    Granderson
    Martin

    I like Gardner because I think he makes an excellent lead-off hitter. Add another 150 Abs to him, project his OBP and increase in SBs and Runs.

    I like Cano batting second to give him the chance Mattingly had with Rickey…more fastballs due to SB threat. Also allows him to get more ABs, especially late in tight games.

    With his new batting stance and IMO projeted power increase, I think Jeter would be great in 3 hole. Lifetime, he has hit better there than any other spot.

    ARod in 4, Tex in 5, and rest is interchangable IMO.

    • Ed

      Brad, I LIKE that order! Nice job!!!! Cash should take a look at this!!!

    • T.O Chris

      I’m fine with Gardner leading off but I don’t see how you are projecting his OBP to go up, if had a pretty monster on base last year and at best I think he would repeat it, I don’t see Brett as a yearly .390/.400 OBP guy.

      I have a problem with Jeter in the 3 hole because even if he does increase his power he won’t be hitting 20 HRs because he hasn’t done that in a while and those days are done (IMO) and I doubt he has a much higher OBP from last year, at least not enough to allow him to hit 3rd with less than average power.

      If we move Jeter out of the top 2 he should go to the bottom of the order.

      I have always been a big fan of batting Cano second though and I would love that to happen because as you pointed out he would increase AB total quite a bit and would see more fastball, however I wonder if he would benefit as much from that as Swisher and Granderson would? Because if hitting 2nd continues to help Swisher hit better (as it did last year) then it may be more valuable to let him hit second since we know Cano should rake anywhere.

  7. Brads

    In looking at the line-up analysis tool, it seems to me that it does not take into account baserunning stats, and the impact of those skills on runs scored. Thus it seems flawed.

    I would love to see it updated to somehow look at ability to steal, go from 1st to 3rd or 2nd to home, and perhaps runs scored as percentage of something other than homers.

    It would also be interesting to see some impact of RISP ability and ability to advance runners…It would be interesting to see how many times each position in order comes to plate with RISP, and figure out where the optimal position is for hitters with high RISP stats or ability to advance runner stats.

    Not taking these factors into account really makes the tool less useful IMO.

    • Agreed, it certainly isnt perfect. That said, the general idea is that speed isnt quite as important a factor as the ability to reach base and the power to drive people in, such that I dont know that it would change the lineups drastically.

  8. Greg Kopp

    I wish I had the last 2 minutes of my life back. These lineups have 0% chance of actually being used.

    • That’s entirely beside the point. The idea was to see what the best lineups would be, statistically. As I said at the end, the lineups that will be used will look much more standard.

Comments are closed.