Wednesday, February 11, 2015

Putting on My Manager Hat



I’ll be honest, I love to complain about batting lineups. In the microcosm of a single baseball game, it always seems like it’s such a huge factor: “if only Harper was up now!” In reality, optimizing your lineup has a very minimal impact on a team’s overall winning expectation. It doesn’t mean Matt Williams shouldn’t put thought into his lineup and it certainly doesn’t mean that old school strategy to constructing a line up is accurate. It really means that there are too many variables in a single baseball game for the lineup to really make a huge difference. But let’s take a closer look at it anyways!

First off, old school thinking about lineup construction isn’t terribly off the mark. In The Book, the authors take a statistical look at what kind of a lineup optimizes the expectation of winning.  Very quickly, a summary: for a lead off player, OBP is most the important factor. The leadoff hitter gets the most at bats in every game, so you want to slot someone here who is most likely to get on base, speed (the old school determining factor) is a nice bonus but not necessary. The second hitter is actually one of the most important, not a “bat control” specialist. He will also get more at bats than others due to his location at the top of the lineup. He is likely to hit with no one on base since he follows the 8 and 9 hitters before the lead off, so OBP is still most important. The third spot is overrated. He hits with no one on base and two outs more often than you think. The cleanup hitter should be just that, the best hitter with power. The fifth hitter gets up with guys on base, so a solid hitter with power is what you are looking for. The rest of the lineup progresses in decreasing ability.

While the lineup in a single game has less impact than luck, there is an impact over a full season. I’ll make a few simplifying assumptions to show you what I mean. If we assume the lineup rolls over 3 ½ times during a single game, the top half of the lineup only gets one more at bat than the bottom. If we assume that same lineup bats every game for a full season, you get some major differences in plate appearances:

Batting Order
PA's (One Game)
PA's (Season)
1
4
648
2
4
648
3
4
648
4
4
648
5
4
648
6
3
486
7
3
486
8
3
486
9
3
486

Of course, this is an oversimplification, but it serves a purpose. The higher the batting position, the more plate appearances. Given the option, would you choose to have your best hitter like Ryan Zimmerman get 200 more at bats during the season? Of course so. This is what frustrated me so much with sticking Harper at the bottom of the lineup last year. Sure, it probably didn’t impact the outcome of a single game, but you are stealing plate appearances over the long term. So while the lineup in a single game doesn’t have a huge effect, it shouldn’t be ignored.

Luckily for the Nationals, they have a lot of flexibility due to the talent of their lineup from 1-8. Injuries can throw a major wrench in that, of course, but Matt Williams has a lot of options. Putting my manager hat on, here is how I go for 2015: for lead off, I’m sure he will pick Denard Span, but I’m not a believer that he can repeat his stellar 2014 campaign. Steamer estimates a .338 OBP, middle of the pack for the team. I’d put Werth back in the leadoff spot. He has consistently put up solid OBP numbers. He doesn’t have top end speed, but that’s not important in the lead off. What he does offer, though, is great baserunning and the ability to take the extra base. I’d keep Rendon in the second spot with his overall hitting ability but not tremendous power. I would put Zimmerman third, assuming he is back and fully healthy. In the cleanup spot I would put Harper since I think he is ready for a break out year and has the most raw power of anyone in the lineup. Fifth is Desmond who doesn’t have the OBP to hit near the top of the lineup but whose power plays well here. Sixth is Span, a bit of an odd choice at the outset, but he would be a table setter for the bottom of the lineup and you would gain some leverage out of his stolen bases here. Seventh is Ramos, to take advantage of Span hitting in front of him. And rounding it off, is Escobar hitting 8th (you could even put the pitcher here, Tony LaRussa style which The Book proves is actually slightly preferred given the right circumstances, and Escobar 9th).

This lineup has the effect of optimizing, in my opinion, the in game win expectancy while also keeping the most talented hitters at the top of the lineup and getting them more at bats over the course of the season. Feel free to disagree, as I’m sure the lineup Matt Williams pencils in every day won’t match to mine.

No comments:

Post a Comment