The Lineup Analysis Tool: 2022 Astros

It’s approaching that time of year when fans like to argue about lineup construction. And (whispering) I really don’t like arguing about batting order lineups. I probably shouldn’t disclose that, because I want you to continue reading the article.

The fact is that the difference in runs/per game between alternative batting orders typically is fairly small. Outside of inexplicably bad lineup decisions (say, batting a really bad hitter at the lead-off position), I don’t feel that it’s worth tremendous effort to debate alternative batting orders. That said, perhaps we can discuss what analytic methods might tell us about an Astros batting order in 2022.

I used the Baseball Musings lineup analysis tool (based on work by Morong, Arnest, and Armbrust) which uses OBP and SLG inputs for each starting player to develop the best batting orders (in terms of expected runs per game). The tool contains algorithms that simulate run-scoring opportunities over the course of a season for each position in the lineup. This depends on the run-scoring environment, which can vary from year to year. Two data periods are available (1959-2004 and 1989-2002) for an average run environment. The first period is a lower average run-scoring period because it includes the low-scoring 1960s. I chose the higher scoring period that begins in 1989, and the average runs/game for all Astros lineups in that environment is 5.286, which is just below the Astros' actual 5.33 R/G in 2021.

I used the ZIPS 2022 projection for Astros hitters as the input for OBP and SLG. I assumed McCormick is in CF, because that seems most likely. The lineup analysis produces a large number of best and worst lineups based on the inputs. Here are the five best batting orders for the 2022 Astros.

(R/G) followed by batting order positions 1 - 9

5.39 Bregman Tucker Gurriel Alvarez Altuve Pena McCormick Maldonado Brantley

5.388 Bregman Tucker Brantley Alvarez Altuve Pena McCormick Maldonado Gurriel

5.388 Bregman Alvarez Gurriel Tucker Altuve Pena McCormick Maldonado Brantley

5.386 Bregman Tucker Gurriel Alvarez Altuve McCormick Pena Maldonado Brantley

5.386 Bregman Alvarez Brantley Tucker Altuve Pena McCormick Maldonado Gurriel

The worst lineups are roughly one-half run per game worse than the best lineups. Most of the “worst lineups” have either Pena at 1 and Maldonado at 2, or Maldonado at 1 and Pena at 2. A lineup with Maldonado at lead-off or No. 2 is highly unlikely. However, if Sunday’s spring lineup is any indication, Dusty Baker appears to be toying with the idea of Pena at leadoff. Would that be a harmful departure from the ideal lineup? That really depends on the accuracy of ZIPS’ projection for Pena (OPS+ of 95). As a rookie, Pena’s hitting stats are particularly uncertain. If Pena beats the ZIPS projection by a good margin, then putting him in the lead-off spot may not hurt very much. But he still isn’t likely the ideal lead-off hitter, given that his OBP is unlikely to be among the team leaders.

It is fascinating that Bregman is the lead-off hitter in all of the highest-run scoring lineups. One of the main reasons is that ZIPS projects Bergman to achieve the highest OBP on the team (.380). In three of the top five lineups, Tucker is the 2 slot hitter. Tom Tango, the sabermetrics expert, contends that the team’s best hitter should be in the 2 slot. Tucker and Alvarez are the team’s two best hitters (in terms of OPS+), according to ZIPS. The other two top-five lineups have Alvarez at No. 2.

Although Maldonado is the worst hitter according to ZIPS, he is at No. 8 rather than No. 9. The top batting orders have good hitters (Brantley or Gurriel) at No. 9. Effectively, the analysis treats the No. 9 hitter as the second lead-off hitter when the batting order turns over.

Any thoughts on the lineups? The batting order debate may not be very important, but get it out of your system now, I guess.