Do lineups even matter?
We certainly think they do. We argue that they do. Oh, and you better believe we will die on a theoretical hill arguing that they do. We all have our ideal lineup – the one we are so sure is better than any manager would use. We have the answer, we think, to make the team better. And the numbers might even support us. Through projections, we can come up with a total we believe our lineup is capable of.
But does it even matter?
Considering all of the wasted digital ink over the years, you’d think it does. And I guess it does, but not nearly as much as we want to believe it does. Lineup construction or optimization is one of those things that seems like it should be a big deal. In the end, though, with most reasonable lineups – the kind where Freddie Freeman isn’t hitting ninth for the Braves – arguing lineup construction is splitting hairs. You might be right, but’s kind of like arguing that PI is actually 3.14159265359 when someone tells you it’s 3.14. You’re both right, but kudos to you – you’re a little more right.
Last year, Zach Kram of The Ringer looked at the evolving leadoff hitter and pointed out this sad truth:
Despite how prominently batting order factors into fans’ view of a manager, and despite the public angst it inspires, the 1-through-9 carries a dark secret: It doesn’t actually matter all that much. Or, as the writer Jack Moore once began a piece at FanGraphs, “When it comes to sabermetric studies, no single item sees more energy expended with less gain than the analysis of batting orders.”
The authors of The Book: Playing the Percentages in Baseball, the 2007 sabermetrics tome that contains the authoritative analysis of lineup order optimization, found that the difference between a perfect lineup and any typically constructed lineup, regardless of its philosophical underpinnings, is worth only about 10 to 15 runs over a full season. At the generally accepted exchange rate of 10 runs per win, that’s not much—and it’s the figure for a full lineup.
But how can this be? How can lineup optimization not have a bigger impact? Here are a few reasons, though I’m sure anyone can add to this list:
1) We’re still talking about the same collection of players. Now, again, I’m not saying that you can randomly put the names in a hat and come up with a lineup of Dansby Swanson followed by Sean Newcomb and then Nick Markakis and you will be just fine. Batting Andrelton Simmons in the leadoff spot despite a sub-.300 OBP as Fredi Gonzalez did far too often in 2013 is still a bad idea. But the effects of hitting Freeman second, third, or fourth and the other moving parts – provided they are sensible decisions – will produce generally very little variance in run production.
2) The plate appearance difference isn’t as large as we think it is. Once again, I’m not saying you don’t want to give your best hitters the most opportunities to bat as possible. That said, in 2018, on average, a team’s leadoff hitter came to the plate 103 more times than the seventh spot in the lineup. Considering their collective .331 on-base percentage, leadoff hitters got on base 34 times with the additional plate appearances. Stretch that over a full season and you’re talking about roughly an additional base hit or walk every week.
3) Managers can be foolish, but most aren’t absolutely idiotic. This goes back to the idea that we’re not talking about extreme lineups. Most managers – even the most traditionalist of traditionalists – wouldn’t bat Freeman lower than third. They’d likely hit Ozzie Albies in the leadoff spot or second spot. They’d probably hit Ender Inciarte no worse than second. When we compare what they would do with the most sabermetric lineup we could possibly come up with, we’re going to be doing a lot of the same things with small differences. Maybe we drop Inciarte, but run production is not rocket science. Whether we believe that a speedy dude who gets on base should bat leadoff or a guy with a minimum of a .340 wOBA should, we’re approaching things with a similar focus.
4) As much time as a manager can spend on crafting the right lineup, how often does it matter where you put this player in relation to this player? In the first inning, it matters. But after that, your cleanup hitter might leadoff innings for the rest of the game. It’s just how the lineup works. So, while you have orchestrated a lineup that crosses off every piece of advice The Book gave you about lineup optimization, you’re spending the entire game hoping your big, bad cleanup hitter can get a rally going.
So, does lineup construction matter at all? Yes. If the completely optimized lineup produces 10-15 runs more than the normal one, that could be the difference in at least one extra win – possibly two. For a team in playoff contention up to the last week, that matters. For a team in the playoffs, you better believe it also matters. In a vacuum – ignoring all other factors – a team should do all they can do to claim an edge.
The problem is this – we actually don’t know what the optimized lineup looks like until we have the data. Then, we can run simulations to find the perfect lineup. Until then, we are guessing. We might be basing our guess on readily available information like projections or recent history, but we could be wrong.
That’s why I stopped really caring about lineup construction. It’s like being upset that Heath Ledger was cast as the Joker because, in my head, I saw 10 Things I Hate About You and A Knight’s Tale. No way can that guy play the Joker. But I was wrong – so very wrong. We don’t know what the 2018 best lineup will look like. We can have our theories and we can cite evidence to support our theory and you know what…we might even be right.
Of course, that tidbit won’t stop us from throwing out the perfect…most AWESOME…lineup possible for the 2018 Braves. And I’ll do that right now.
1. Ozzie Albies, 2B
2. Freddie Freeman, 1B
3. Ronald Acuna Jr., LF
4. Tyler Flowers/Kurt Suzuki (a.k.a. Flowzuki), C
5. Nick Markakis, RF
6. Johan Camargo, 3B
7. Dansby Swanson, SS
8. Pitcher’s Spot
9. Ender Inciarte, CF
This lineup isn’t entirely sabermetric. Inciarte wouldn’t have dropped to ninth in that kind of order. In fact, he likely would have hit third or fifth. But I prefer my idea. I want to give Freeman as many opportunities to hit because he’s my best hitter and I also want baserunners in front of him. After the first inning, I envision Inciarte and Albies beginning rallies for Freeman to clean up. It’s the best of both worlds.
Once established, what Freeman doesn’t clean up, Acuna Jr. can. The rest of the lineup is a bit of a mystery, but I like the idea of keeping Inciarte/Albies/Freeman/Acuna Jr. in a pack without sacrificing plate appearances so that Freeman hits third or fourth.
Feel free to post your ideal lineup below. While doing it, also accept it’s just for fun because short of you inverting the lineup I have, the run production of our different lineups won’t be significant.