Detroit Tigers 2021 ZiPS Projections: Good and Bad

What the ZiPS projections get right and wrong about the 2021 Detroit Tigers.

Earlier this week FanGraphs released their ZiPS projections for the 2021 Detroit Tigers, and as you may surmise, the team looks like it’s going to be quite bad.

If you’re unfamiliar with ZiPS, it’s a computer projection system created by Dan Szymborski that went live in 2004, and each year it offers estimates of MLB production for just about every interesting player in pro baseball. It’s probably best to let Dan describe what it does himself:

"ZiPS moved quickly from its original inception as a fairly simple projection system: It now does a lot more and uses a lot more data than I ever envisioned 20 years ago. At its core, however, it’s still doing two basic tasks: estimating what the baseline expectation for a player is at the moment I hit the button; and then estimating where that player may be going using large cohorts of relatively similar players."

Dan is a great guy. He’s a fun Twitter presence, he joined Rogelio and me on a podcast last year to talk about the 2020 projections, and I often reference ZiPS in my pieces on trades and free agents. By no means am I trying to discredit ZiPS in any way, I just want to take a close look at what seems right and what seems wrong in this year’s projections.

What ZiPS Gets Right

Jeimer Candelario

Maybe this is wishful thinking on our part, but Jeimer Candelario’s breakout season last year seemed very real, and largely sustainable, minus 20-30 points in batting average. ZiPS projects him to hit .252 with 54 extra-base hits and above-average defense. He struggled a bit with the glove at first base, but those numbers all seem perfectly reasonable for Candelario next year.

Isaac Paredes

In case you were wondering about playing time projections, Szymborski writes “ZiPS is agnostic about future playing time by design.” So no, I don’t think Isaac Paredes will get 530 plate appearances in 2021, but that’s not what ZiPS is trying to predict. It’s trying to predict what Paredes would do with that many plate appearances, and I think these numbers are very reasonable. If anything, I’d suggest the power numbers are a bit high, given Paredes’ struggles to hit the ball hard last year. But Detroit’s young infielder typically started slowly in the minors, but he eventually shows the ability to adjust to his level of competition.

Matthew Boyd

There’s one stat in the above projection that could completely derail what looks like an otherwise reasonable projection: HR/9. Matthew Boyd owns the 11th best strikeout-to-walk ratio in baseball over the past two seasons, but all that good work has been ruined by extreme home run issues. His 2.24 HR/9 last year was the highest rate for any qualified starter in Major League Baseball history. His 1.89 HR/9 in 2019 was the 17th highest ever.

So why do these numbers seem right to me? Because Boyd’s stats are truly unprecedented, and I have to believe he’ll find a way to fix his home run issues. And it’s not as if a 1.5 HR/9 is some impressive number — it would have been the 6th worst rate among starters last year — but it at least seems plausible.

What ZiPS Gets Wrong

Spencer Turnbull

Spencer Turnbull’s projection looks pretty reasonable, but as with Boyd, ZiPS appears to be regressing to the mean on home runs. That just doesn’t make sense with Turnbull.

Spencer Turnbull has proven to be elite at avoiding home runs. It’s possible his home run rate spikes in 2021, but it’s hard to envision it doubling for an entire season. If it stays at his career rate, and the rest of the ZiPS projections stay the same, Turnbull’s ERA drops from 4.51 to 3.54. Even if his home run rate jumps to 1.0 HR/9, his ERA drops to 4.00. I can absolutely see Spencer Turnbull posting a 4.51 ERA, but I don’t think it will be because of home runs.

Gregory Soto

Relievers are notoriously volatile, so it wouldn’t be a a huge shock if Soto puts up these numbers, but this projection is still a bit strange. Gregory Soto can be wild, but he hasn’t posted a walk rate that high since he was in rookie ball in 2015, and ZiPS is also predicting a big spike in Soto’s home run rate and an 8% drop in his strikeout rate. That sounds more like a worst-case scenario than an average outcome, so I’m betting Soto beats this projection by a decent margin.

Matt Manning

This just feels far too optimistic to me. That isn’t to say Matt Manning is bad prospect by any means, it just seems highly unlikely he would end up more effective than Casey Mize and Tarik Skubal in 2021, let alone finish with the 3rd best WAR and lowest ERA among the team’s starters.

Manning absolutely has the talent to put up these numbers one day, but this seems like a case of minor-league stats deceiving ZiPS. Manning simply overpowers minor-league hitters with two plus pitches and above-average control, but that’s not quite enough to succeed in the big leagues. He’ll need to take significant steps forward with both his changeup and his command to find consistent success at the highest level, and a 4.43 ERA in 2021 seems about a run too low.