Replacing Torii in the Two-Hole (Part 3) – Creating the Most Prolific Lineup

By Gregory Bird – AngelsWin.com Columnist

In our first two articles we looked in depth at Torii’s performance last year batting second. We also examined the candidates to replace him. But it’s still unclear, so we ask again: Who goes in the two-hole in 2013? Of course, we all want a lineup—billed as the most prolific of our era— to live up to expectations and carry the Angels to a World Series victory. Some of what I written about here may be controversial and it may fly in the face of all we thought we knew. This wasn’t the conclusion I intended to write when I started this project. Hopefully, we can all keep an open mind. If we do remain open we may be able to challenge our old ideas and have a great discussion about scoring runs.

There is a great article over at SB Nation by Sky Kallman, “Optimizing Your Lineup by The Book,” that summarizes research by Tom Tango, Mitchel Lichtman, and Andy Dolphin in their book, The Book. And, in short, on-base percentage (OBP) is the most important statistic in determining a lineup. Slugging percentage is a close second. Other stats like platoon splits and base stealing ability can be used to tweak the lineup but are of minor importance. One of the biggest pieces of information is that optimizing production is not accomplished with simple baseball logic. For example—and surprisingly—hitting your best player third, ordinary baseball practice, does not optimize run production at all, because more often than other hitters he bats with two outs and nobody on base.

In constructing a lineup, we should remember that each spot higher in an order gets more at bats. To maximize your better hitters, in terms of OBP, the best hitters (with regard to avoiding outs) should bat higher in the lineup. In addition to simply minimizing outs over a season it is important to put the right hitters in higher leverage situations. A high leverage situation is, for example, when someone is in scoring position with two outs. The Book’s research suggests that a team’s three best hitters, in terms of OBP, should bat first, fourth, and second (in that order.) 

The leadoff hitter is optimized with the highest OBP and the least power (slugging percentage,) since he comes to bat with the bases empty 8% less than any other hitter in the lineup. The leadoff hitter comes to bat with a guy on base only 36% of the time. The four-hole hitter comes up to bat in the most important and high leverage situations. For example, think about a first inning AB for aA four-hole hitter: for this to happen a guy must be on base and there is most likely an out or two. This is a key moment for a potential run to be pushed across. The two-hole hitter is very important as well. He is much like what the traditional three-hole hitter has been assumed to be, but with more at bats. He needs some pop to drive in the leadoff hitter, who is likely on base when the two-hole hitter comes to the plate. He also needs a good OBP of his own to maximize the additional at bats he gains and to keep the lineup moving. I realize all of this is counter to the generally accepted wisdom of baseball, which is included in this article if you read it. As I looked around more I found this same research repeated by others and others coming up with some of the same conclusions.

But what about steals, you may ask. Isn’t it important to have a fast leadoff hitter who, for the big hitters behind him, can steal second and get into scoring position? I want to take a moment to answer these questions which I’m sure a few of you are asking. There is research on it and it was surprising to me, too.

The research says if you want to maximize a base stealer than he should hit in front of a singles hitter. Base stealing isn’t really maximized in front of power hitters. The conclusions were that unless your base stealer deserved to be in one of the top three spots in the lineup (first, fourth, or second) then he should be placed in the middle of the lineup, just before the slap hitters. And it makes sense: a guy who steals bases is going to get caught from time to time, erasing himself from the basepaths. If he is caught just before a power guy hits a homerun then it is a waste and his stolen base attempt, even if he would’ve been successful, was also wasted. He would’ve scored no matter what if he would’ve just stayed put.

Additionally, if a guy steals second he is going to score on almost any hit except an infield one. What value does a stolen base provide if the next hitter is going to hit a double or a homerun? It is best not to risk getting erased from the bases in order to maximize the next hitters RBI total and the total amount of runs scored.

Now if the guy behind you is most likely going to only hit a single, then there is great value in being on second base for his AB instead of on first. This research doesn’t say it isn’t helpful to steal bases in front of power guys but it does say that, all else being equal, the value is maximized in front of singles hitters. Take it for what you will. I’m sure a few of you may have some push-back to voice, but think it through first.

Where does this leave us?

Our three best hitters in terms of avoiding outs are Mike Trout, Albert Pujols, Josh Hamilton; Chris Iannetta is a close fourth to Josh deserving honorable mention here. This means that one of our big three needs to hit second in order to maximize our run production. What?! I know this seemed crazy to me too. Of the three, Trout and Pujols have been the best at avoiding outs so they are the best candidates. 

I looked around for a way to analyze this. I found a lineup analyzer tool over on baseballmusings.com and since it was all I had access to I used it. I wanted to test this information out as best I could so I did the research. The tool is based on a lot of sabermetric scholarship and isn’t perfect but it did the trick. 

But before I start comparing 2013 lineups using this tool I wanted to see how the tool measured up. I tested last year’s lineup to actual results. Last year the Angels scored 767 runs which placed them fourth in MLB for runs scored. 

First problem in testing this tool is that Scioscia used a ridiculous number of lineups. I settled on three different ones to use to test it out. The ‘pre-Trout’ lineup was used for the first 20 games and looked something like this: Aybar, Kendrick, Pujols, Morales, Hunter, Wells, Callaspo, Iannetta, Bourjos. Second was the ‘post-Trout’ lineup which was used for the next 38 games and it looked something like this: Trout, Callaspo/Izturis, Pujols, Morales, Trumbo, Kendrick, Wells, Aybar, Wilson. I used Callaspo’s numbers during this period over Izturis’ because frankly his were a lot better. Now finally we have the ‘Hunter in the two-hole’ lineup for the remaining 104 games which was: Trout, Hunter, Pujols, Morales, Trumbo, Callaspo, Kendrick, Aybar, Iannetta.

I put the numbers into the machine and here’s what I got. The ‘pre-Trout’ lineup should’ve produced about 3.2 runs per game or 64 runs over 20 games. The ‘post-Trout’ lineup produced about 4.693 runs per game or 178 runs over 38 games. The ‘Hunter in the two-hole’ lineup should’ve produced about 5.255 runs per game or 547 runs over 104 games. These three total up to 789 runs or 22 runs over the actual amount scored. This discrepancy could easily be explained with all the Scioscia line-up experimentation throughout the season that I can’t really reproduce here. All in all I felt the analyzer was fairly accurate and passed the test.

I decided to evaluate four different statistical sets for Angels’ players; their last full season stats, their career stats, Bill James’ liberal predictions from FanGraphs, and the über conservative ZiPs projections. As we all know, based on Scioscia’s lineup creativity, we will not have the exact same lineup all year. So, I decided ballpark figures were good enough for our discussion and here are the six lineups I tested. 

The first three lineups are what I would imagine Scioscia would use to start the season. These are the lineups with our three original candidates and how I think those lineups will shake out one thru nine. The fourth and fifth line-ups are using information from the data provided by The Book. The fourth one is strictly by The Book while the fifth one is merely influenced by it. The sixth lineup is influenced by what I learned from the line-up optimization data and sprinkled with the realities of current major league lineup construction. You can see in the fifth and sixth line-ups I’m placing Iannetta at leadoff.

Iannetta, while not our best on base guy, does have many of the qualities of a leadoff hitter. He is patient, takes lots of pitches, gets on base well, and doesn’t have a lot of power. He is only slightly less effective in his career, by only 9 points, at getting on base than Hamilton and could easily replace Josh in the group of ‘top three best hitters’ at avoiding an out. The only thing that is bad about Iannetta leading off is his speed. He calls himself a base clogger. He is below league average at scoring when on base and in taking extra bases but does that really matter? The good news is, no. The guys behind him will be crushing the ball. Iannetta won’t need to have incredible speed to get himself to third on a double or home on a homerun. The data suggests he doesn’t need to be great at taking extra bases to lead off. He could end up limiting Trout’s base stealing a bit but it may not be that big of a deal as he could also increase Trout’s RBI total. We have our lineups so let’s see how they all tested out:

The very first thing I noticed about this information is something I should’ve paid attention to when I began this whole process, the differences aren’t that big. Early on I came across this quote, “Believe it or not, the difference between an optimized lineup and a typical, mildly foolish one you’ll see MLB teams use is only about one win over 162 games.” I pressed ahead anyway and the results of my data prove this point. The differences are just not that great, only 14 runs over a whole season at most.

Has this all been a waste then?

No, because we now know we don’t have to argue so much over lineup construction. It just isn’t going to affect the outcome often. Second, we can still argue over how best to set the lineup up for the one extra win we could need come September. Finally, look at those season run totals! In the most conservative estimates we should score somewhere between 30 and 40 more runs than last year. If we look at career numbers we should score between 100 and 130 runs more than last year. This would come close to matching our season run totals from 2009 (883R) when we won 100 games. Notice also that we should average somewhere around 5 runs per game. This should mean a lot of wins for Victor and Gubi to call in 2013. That to me is great news and well worth my effort.

I know this is anti-climactic after three articles and a ton of information but here are my final thoughts on who should hit in the two-hole. I think it should eventually be Trout. If not in 2013 then I’d like to see Trout get moved there during his career because he could be a prolific run producer with his power. I would love to see Bourjos or Aybar get their OBP high enough to make them viable leadoff guys but I just don’t think it’ll happen this year for Bourjos or ever for Aybar. Until this happens I like Iannetta and his 80 plus walks per full season leading off. We can maximize his walks and most effectively utilize his OBP by batting him first. Remember, less outs in the top of the lineup means more runs and more wins. 

Looking at the data, if Trout isn’t batting second then it doesn’t really matter to me who hits there. The difference over the whole season of any one of the three original candidates is 3-8 runs. I would expect Scioscia to select the Aybar 2nd lineup and in the end I’ll be ok with that. The bottom line is we are going to win games. The lineup should not be our problem, no matter who bats in the two-hole.

The real question is the bullpen and will it blow leads? None of us know how that will turn out. Even if the bullpen does blow saves our lineup will be able to take some back with dramatic walk-off hits in the 9th. I hope 2013 is as fun as I believe it will be, as fun as it looks on paper, but we all know the dread of that statement. I’m sure a lot of you have an opinion on this and I’m excited to hear it so please leave your thoughts about the lineup. In the meantime let’s sit back, relax, and enjoy the ride that will be the 2013 regular season.