Three Twins Takeaways from ZIPS and Steamer Projections

[Matthew Lenz]

We now have ZIPS and Steamer projections for 2020. After (obviously) looking at the projections of total home runs where Steamer has the Twins at 262 and ZIPS at 264, the attention was then turned to look for notably different projections.I started by sorting both groupings by WAR and noticed right away that Randy Dobnak was projected quite a bit differently. ZIPS has him projected for a WAR of 2.1 whereas Steamer is at just 0.7. For reference, Dobnak contributed a 0.8 fWAR in 2019 in just 28 1/3 innings pitched. Comparing K/9, BB/9, HR/9, and ERA it’s clear that Steamer doesn’t quite buy into his 2019 campaign like ZIPS does, although the biggest explanation for this is innings pitched. Steamer has him pitching just 90 innings in 2020 whereas ZIPS has him at 147 innings. Regardless, a 0.7 fWAR over 90.0 innings is quite a bit different than 0.8 fWAR in 28 1/3 innings. It would be fair to temper expectations for the Twins 2019 Minor League Pitcher of the Year, as he only gets to 147 innings due to injuries from the starting staff.

 

Many Twins fans are awaiting the true breakout of Jose Berrios and view 2020 as the year it will happen. After a good first half in 2019 he struggled to carry that into the second half of the season. Unfortunately, both ZIPS and Steamers view him as a “quality” starter in 2020 which isn’t what you want to hear about your number one. Quite literally, both projections have him average six innings per start and giving up close to three runs, although ZIPS is slightly more optimistic on him which turned out to be a theme between the two projections. Essentially, both projections view him as a slightly better version of Kyle Gibson’s career as a Twin which is another thing you don’t want to hear about your number one.

 

On the offensive side, Twins utility man and third catcher Willians Astudillo is projected quite a bit differently. ZIPS has him contributing 1.5 WAR whereas Steamer has him at just a 0.3 WAR. This is largely due to Steamer projecting him for just 50 at-bats, which is realistic considering the signing of Alex Avila and 2019 emergence of both Jake Cave and Ehire Adrianza as quality role players. Like Dobank, at this point in his career, Astudillo’s comical character may overshadow his flaws as a contributor at the major league level.

 

MORE FROM TWINS DAILY

— Latest Twins coverage from our writers

— Recent Twins discussion in our forums

— Follow Twins Daily via Twitter, Facebook or email

 

Click here to view the article

[Richard Swerdlick]

Projections, projections. In the Fall of 2016 most projections had Hillary beating Trump too. Say hello to President Hillary Clinton. So much for statistical projections.

[Cody Pirkl]

With regards to Berrios, Steamer makes absolutely no sense to me. They're expecting a decrease in K/9, increase in BB/9, a ridiculous spike in homers despite coming off a 1.17 HR/9 in the year of the juice ball, and basically the worst year of his career as a whole since his debut. It looks like Steamer somehow thinks Berrios has peaked at 25. As somebody who looks around at projections quite a bit for fantasy baseball, Berrios has one of the most absurd projections I've seen for seemingly no reason.

[vavo]

Projection systems typically do a poor job highlighting the likely range of outcomes. You can understand why, since a projection that says, "We think there's a 90% chance Miguel Sano hits between 10 and 55 HRs." is kind of a boring thing to say. It's more satisfying to make specific claims, like 35 HRs. But similar to the political projections Richard mentioned, they're actually calculating a range of outcomes. 538 gave Clinton a 2/3 chance of winning and Trump a 1/3 chance of winning. They weren't "right," but they weren't really wrong either. If a projection system said Berrios has a 1/3 chance of having an ERA under 3, or Arraez has a 1/3 chance of hitting above .350, or the Twins have a 1/3 chance of making the World Series, I'd be pretty excited!

 

It's too difficult to effectively show the range of outcomes on everything, but I'd like to know those sorts of things. Even if the middle of their projection has Berrios as being worth 3.1 WAR, what sort of chances does ZIPS give him of a huge breakout, say being worth 6 WAR or more?

[joefish]

A sphincter says WHAT?

[jimbo92107]

I think Dobbs is a stud horse in the making. 147 innings is his low end. The man is built to chug along like a diesel engine. Easily could go 170 or more this season, then top out at over 200 innings for several seasons. 

[108Stitches]

Steamer is a joke. They have Tyler Wells with an ERA nearing 6 every year and he’s got a career 2.82. They only care about draft position

[Mr. Brooks]

Steamer is a joke. They have Tyler Wells with an ERA nearing 6 every year and he’s got a career 2.82. They only care about draft position

Take a look at the innings pitched.

They aren't actually projecting an ERA for Wells (or any of the other pitchers they have at 1 inning).

It just means they don't project him to pitch in the MLB and they don't want any of those guys that won't pitch to list above the players who will when sorting by ERA.

It's not a knock against Wells, it's the same for everyone they don't expect to play, regardless of draft position.

[Muss0070]

I think this article would be more well written if you included:

1. Why you think one projection is more accurate than the other.

2. Your projection.

3. Is WAR as your baseline to determine deviations the best way of approaching this? I am not sold on it personally.

4. Analysis on players that may have had "fluke" years. (I.e. arraeza, Taylor rogers, duffey, Pineda, Buxton, sano)