In case you missed it, FanGraphs posted the ZiPS projections for the Cleveland Guardians for 2023. Also, in case you missed it, those ZiPS projections are compiled by FanGraphs’ writer, Dan Szymborski, who is an excellent baseball writer as well as one of the top baseball data analysts as the creator of ZiPS.
If you’re not familiar with how ZiPS works, I recommend checking out Dan’s explanation of the process he goes through to produce the projections and his insights into how to interpret the results properly. Additionally, MLB provides a glossary entry with a helpful definition of ZiPS.
Dan was kind enough to agree to answer some questions I had after reviewing what ZiPS has to say about the Cleveland Guardians for the upcoming season.
[Responses have been lightly edited for clarity]
Quincy: Dan, first of all, thanks for the fun tool of ZiPS for us to have something to talk about in the long offseason and thank you so much for being willing to answer some questions about the projections just released for the Cleveland Guardians.
My first question revolves around the shift … was ZiPS asked to adjust for potential changes in outcomes for hitters, pitchers or defenders with the shift rule changes, or is that something that is somewhat impossible to predict going into this first season under the new rules? Guardians fans have speculated that players like Josh Naylor and Josh Bell may benefit as players who hit line drives and often face a shift while batting left-handed, is this kind of thinking faulty from your understanding?
Dan: Everything I do in ZiPS is something that I can test and base on data that exists. New rules is kind of guesswork and the results of such changes aren’t always obvious. And with something like the shift, I don’t think we really have anything to compare to. I don’t believe teams are going back to the traditional defensive positions; I expect them to shift as much as possible, with a fielder just on the left side of second base, and that’s kind of a “mostly-shift” configuration that we don’t have the direct data for.
Quincy: In a related question, I think many Guardians’ fans read your suggestion to move Giménez to shortstop and nodded in vigorous agreement. However, a recent article from Travis Sawchik suggested that second-base defense might be more important than ever because of shift changes. Do you think it’s possible that Giménez playing second base might be an attempt by the Guardians to adjust to changes they expect to take place in terms of defensive opportunities for second basemen?
Dan: I’m not convinced that second base becomes more important. After all, we have a lot of information from before shifts became as popular as they are now, and the data, both direct and implied, suggested that shortstop was the more crucial position and the one that teams moved players from to an easier one.
Quincy: Are there general principles we should keep in mind when comparing projections from Steamer and ZiPS? For example, as a Guardians fan, I want to run with Oscar Gonzalez’s and Trevor Stephan’s numbers from Steamer and Mike Zunino’s and Triston McKenzie’s numbers from ZiPS. Just curious if you’ve seen any underlying reasons over the years as to why some of these projections can be so different between the two systems.
Dan: This one is really hard to say simply because while I know what goes into ZiPS, I cannot say for sure what goes into Steamer. I think generally you should look at as many good systems as you can, but I’m not sure you’ll find systemic differences everywhere; accuracy is tricky, but we can deal with bias for certain types of players, and I’m pretty sure most of the systems have picked that low-hanging fruit.
Quincy: I was very excited in 2022 to see George Valera projected as one of the top eight hitters in baseball for wOBA in 2024. With the 2023 projections displaying some fun comps for Valera at his age to Dwight Evans and Aaron Judge, it seems I should still feel optimistic for his future. Can you share any further insight on what ZiPS sees in Valera and if you see him as someone the Guardians should be looking at as an option in right field at some point in 2023?
Dan: I think it’s important to remember that Dwight Evans and Aaron Judge weren’t Dewey or Arson Judge at that point. It’s comparing baseline and since individual comps aren’t really that important, it’s mostly fun and a reminder that a player of Valera’s baseline has developed into a damn good hitter at some point.
Quincy: In discussions on projections over the years, I have often heard Guardians fans say that projections systems can tend to underestimate hitters (like Steven Kwan) and pitchers (like Cal Quantrill) who are somewhat BABIP reliant. In your experience, can there be something in a player’s hit/pitch profile that would make them more likely to regularly beat their xwOBA or xFIP like Kwan and Quantrill, respectively, did in 2022, or is that a phenomenon that should never be projected as repeatable?
Dan: I haven’t found any pattern as such. That’s the kind of thing tools like principal component analysis are able to catch when looking over the data. If ZiPS had an underperformance in this manner, it’s built to take that into consideration. It doesn’t rely on peripheral-based numbers, it uses them along with real numbers, based on the historical predictive value of those numbers. Now, it’s possible there’s some subset that I haven’t had the imagination to look for, but calibrating the relationship between actual numbers and peripheral-type numbers in predicting the future is one of the things I look at the most heavily.
Check out Dan’s work on fangraphs.com, and please consider subscribing to the site to support its writers and the wealth of useful content the site produces for baseball fans. Also, you can follow Dan on Twitter at @DSzymborski.