clock menu more-arrow no yes mobile

Filed under:

Optimize your own (damn) (Braves) lineup

Or just pick names out of a hat.

MLB: Philadelphia Phillies at Atlanta Braves Brett Davis-USA TODAY Sports

The baseball zeitgeist in April is always a little strange. There’s all this euphoria about baseball coming back, and that leads to this almost-nervous energy about trying to categorize everything we’re seeing as “real” or “not real.” Not “not real” as in, “this didn’t actually happen,” but rather, “this is unlikely to continue.” But, all we have right now are small sample sizes. Our lizard brains wants to latch on to something; it’s instinctual.

Right now, the Braves-fandom-zeitgeist, if such a thing exists, seems to be partially focused on Ender Inciarte. Not just on the man and his offensive struggles to date (to which, really, it’s been two weeks in April), but on what should be done with the lineup. Brad wrote some thoughts on this situation earlier this week, which I agree with: it’s not so much (or at all) that Ender Inciarte batting leadoff is a problem because he’s struggled for two weeks. It’s really more that he’s not necessarily an ideal leadoff hitter anyway, because he’s basically an average hitter (career 94 wRC+, finished each of the last three seasons between 97 and 100 wRC+), and you really want your leadoff guy to be one of your best hitters.

Of course, that’s where the Braves run into somewhat of an issue, and partially explains why Inciarte has indeed been batting leadoff for almost his entire Braves tenure so far (he hit leadoff all of last year, but in 2016, he hit second and even down the lineup at 6, 7, 8, or even 9 a fair bit). In short: the Braves don’t actually have many better options to bat against right-handed pitchers, in general. At least, not yet, or not right now. Sure, there’s Freddie Freeman, but other than that, who else ya got? Nick Markakis is also an average bat. Ronald Acuña isn’t up yet (maybe that’ll change by the time this is published). Tyler Flowers and Kurt Suzuki may be options. Preston Tucker and Ozzie Albies, if you want to put a whole lot of stock into some small samples, perhaps? It’s a tough call. There’s Freeman, and then there are some serious question marks, or guys that aren’t necessarily better than Inciarte at batting against righties.


I’m going to change gears a bit and talk about lineups more broadly. I think lineups get people bent out of shape not because they’re really all too important, but because:

  1. They’re really easy to fix. There’s no real pressure to putting a different lineup out there, it’s not a heat-of-the-moment decision. And, the research already exists to give you a better sense of how one lineup might work better than another, even if the actual difference is small. The easier something is to improve, the more annoying it is when it isn’t improved.
  2. They play into our basic tendencies to focus on criticizing things that go wrong, rather than doing that but also praising things that go right. When was the last time you thought to yourself, “Man, this well-constructed lineup worked like a charm and cut through the opposing pitching like a hot knife through butter?” Probably never, unless you think to yourself in tired similes. But, you get what I mean. It’s really annoying when a worse hitter batting towards the front of the lineup makes an out. But when the opposite happens and the team scores runs, I don’t think anyone ever thinks, “Yeah, and it’s thanks to this specific arrangement of players in this order that that happened.” Because it isn’t.

And that’s why lineups don’t matter very much. Order schmorder, the players still need to avoid outs. But, even though they don’t matter, that doesn’t change the first point — it’s super-irritating when they’re not optimized, because they can be.

As far as lineup optimization itself goes, I’m increasingly of the opinion that the difficulty isn’t in understanding an optimized lineup. By that, I mean — we already have a decent sense of what such an optimized lineup would look like. This excellent feature by Sky Kalkman gets referenced a ton, and for good reason. I’m going to do so again, and I’m not really going to discuss this much more. If you want to argue, we can do so; if you just want to learn, I suggest reading it or picking up a copy of The Book. But, here are the basic “optimized lineup” principles.

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.

(There are also various lineup simulators and other software/statistical packages that can examine this sort of thing and probably yield similar-esque conclusions. If you think you can improve on these general ideas in a demonstrable way, I highly encourage you to do so!)

So, anyway, if we have some basic tenets of “this is an optimized-ish lineup” that we agree on, what’s the issue? Well, the issue is the same issue we always have in baseball — we don’t really know what’s going to happen. Sure, it makes sense to bat “your best hitter with power fourth.” But, how do you really know who that is? Player performance is variable, and we have oodles of projection systems, but none of them are perfect or prescient. Sometimes, it’s clear that Player A is much better than Player B at something. Other times it isn’t.

This is all a long-winded way of getting to what I’ve embedded below — a tool to let you optimize your own (damn) Braves lineup to your heart’s content. To do so, here are the inputs it needs from you:

  1. The handedness of the opposing pitcher;
  2. What projection system(s) you want to use to forecast player performance;
  3. The actual arrangement of players you’re going to use; and
  4. [optional] Any other edits you want to make to the forecasted stats.

The actual calculations are simple. The math is basically:

  • First, look at the eight players. Pick the best overall hitter, and bat that guy second.
  • Then, take the two next-best hitters. Find out which one is relatively better at OBPing, by comparing his slugging skill (on a percentile basis) versus his OBP skill (on a percentile basis). The one better at OBPing bats first, the one better at slugging bats fourth.
  • Then, the remaining spots are filled out quality-wise, by going #5, #3, #6, #7, #8.

That’s it. You can play with it, below.

While hopefully intuitive, there are dropdowns for pitcher handedness, the projection system to be used, and then for your choice of player at each position. If you want to tinker further with assumptions, you can just go into the “lookup” sheet and edit to your heart’s content.

So, there you go. Now you’re armed with a very rudimentary tool to optimize lineups. Before I let you go, here’s what this doesn’t do:

  • It doesn’t adjust for an alternation of lefty and righty hitters. This is important, perhaps more important than optimization itself. Unfortunately, most of the good-or-even-average Braves hitters are lefties, though this should change when Flowers returns and Acuña is promoted. As a result, this tool is going to spit out some very lefty-heavy tops of the order, which may not be ideal.
  • For platoon splits, this tool uses historical career platoon splits, except where they are egregious, unsustainable, and/or the product of a small sample, whereby it uses leaguewide platoon splits instead. Again, if you disagree, you can just edit the “platoon splits” columns in the “Lookup” sheet.

With no editing, here are the different lineups it spits out, using the arrangement we’ve seen so far as the “default lineup” with Flowers out:

  • Against righties: Markakis, Freeman, Tucker, Inciarte, Albies, Flaherty, Suzuki, Swanson
  • Against righties: Inciarte, Freeman, Albies, Tucker, Markakis, Flaherty, Swanson, Suzuki
  • Against righties: Markakis, Freeman, Tucker, Inciarte, Albies, Suzuki, Swanson, Flaherty (this is also the “average” such lineup)
  • Against lefties: Swanson, Freeman, Markakis, Suzuki, Albies, Inciarte, Tucker, Flaherty
  • Against lefties: Swanson, Freeman, Tucker, Suzuki, Albies, Inciarte, Markakis, Flaherty
  • Against lefties: Swanson, Freeman, Markakis, Suzuki, Albies, Tucker, Inciarte, Flaherty
  • Against lefties: Swanson, Freeman, Tucker, Suzuki, Albies, Markakis, Inciarte, Flaherty (this is the “average” such lineup).

Anyway, all of these are frankly a little ridiculous, but it’s a fun little exercise. You can adjust the inputs to your liking to make them less ridiculous, I suppose.

My main takeaway is that the Braves have kind of an annoying lineup situation right now. There are too many same-y hitters, and the less same-y guys (Albies, Tucker) are prone to some sample size issues such that it’s hard to know exactly what they’ll do. My real conclusion is that the lineup really needs Acuña and Flowers back.

Sign up for the newsletter Sign up for the Battery Power Daily Roundup newsletter!

A daily roundup of Atlanta Braves news from Battery Power