Dan Szymborski released his annual statistical projections (on Fangraphs) for the Twins roster earlier this week. Called ZiPS (SZymborski Projection System), these are generated by a computer program and are widely considered to be among the best projections in baseball. Let’s take a look at what the almighty Computers think of the Twins roster.

Batters

ZiPS predicts the Twins roster to be aggressively average as far as position players go this year. Almost the entirety of the projected roster (for batters) fall between .7 and 2.2 WAR, with Ehire Adrianza (.3) and Nelson Cruz (2.6) being outliers. However, there are a few interesting projections I’d like to point out:

• Eddie Rosario is projected to have a WAR of 2.2. He produced to a tune of 3.4 fWAR last year, and is just entering his prime, so I’m a bit more optimistic than the computers on this one. One factor that could have contributed to the predicted regression was his relatively high .316 batting average on balls in play (BABIP), although his career BABIP is fairly in-line with this.
• There has been much talk lately of Max Kepler being primed for a breakout this year, but ZiPS is predicting more of the same from Kepler. He’s projected to drop from 2.6 fWAR last year to 2.2 this year, although his batting average is projected to rise from .224 to .246. This jump is definitely due in part to his abysmal BABIP last year (.236).
• ZiPS doesn’t love Byron Buxton or Miguel Sano. At 1.9 and 1.7 WAR, respectively, the computers do not project big bounce back seasons for the former top prospects. However, the system does have trouble properly valuing injuries, so hopefully we can expect more from these two. Buxton’s 2017 fWAR was 3.5, and he draws a lot of his value from his defense, which is not a question mark if he stays healthy.
• It can be interesting to think about possible roster construction through a ZiPS lens. Jake Cave is projected to put out 1.0 WAR this year, while Tyler Austin is projected to be worth .7 WAR. On top of that, Austin is projected to be essentially the same player as Lucas Duda, both having a .7 WAR and eerily similar stat projections across the board.
• The Computers do seem to have the same favorite player as everyone else around here. Yes, I am talking about El Tortuga, Willians Astudillo. ZiPS projects him to have the 4th-highest position player WAR on the team, at 2.0. Last year, he was projected to have a WAR of .6, and his fWAR indeed ended up at .6, so here’s hoping.

Pitchers

The Twins pitchers are (surprise!) projected to be largely average this year by ZiPS. The projected pitchers in the starting rotation (along with Kohl Stewart, Zack Littell, Adalberto Mejia, Lewis Thorpe, and Stephen Gonsalves) all fall between 1.0 and 2.0 WAR, besides Jose Berrios. We have essentially nine number three or four starters (nothing new here). Let’s run through some of the more interesting projections again:

• While Berrios is projected to have the same WAR as last year (3.2), he is projected to regress in areas such as innings pitched, K/9, and ERA. Taking into account his strong work ethic and desire to succeed, I’d like to think that he will instead outperform his 2018 season.
• Kyle Gibson has not yet proved to ZiPS that Good Gibby is here to stay. He’s projected to fall from 2.9 fWAR to only 2.0, due to a predicted regression from his career bests (ERA, K/9, IP, and BABIP against) last year.
• Michael Pineda, coming off a lost season due to injury, is projected to put up only 1.0 WAR. This value would be the lowest in his career by .4 WAR, besides his injury-shortened 2017 season (.8 WAR through half of July). This could be another case where ZiPS projections are too pessimistic due to past injuries.
• Fernando Romero is projected to be our most valuable reliever (1.3 WAR), although this probably does not take into account his forthcoming move to the bullpen. Aside from Romero, the most valuable relievers are Taylor Rogers (1.1) and Andrew Vasquez (.8), although Trevor May’s K/9 projection is encouraging (9.74). If May were able to post that many strikeouts, he would make for a good high-leverage reliever or closer.

You can find the rest of the Twins ZiPS projections here. All of the stats for this article were pulled from Fangraphs.com as well.

What do you guys think? Am I too optimistic? Are the Computers too pessimistic?