Cleveland Indians and Projection Systems: Position Players

(Chuck Crow/The Plain Dealer)

As one begins to prepare for the Indians reporting to spring training, a mere 38 days away, position depth and overall expectations regarding the upcoming season’s roster become increasingly important. Perhaps the Indians have stunning trade left in them but the reality is that limited financial flexibility intimated by the organization suggests that what is currently on the roster is likely what fans will see when opening day comes. With this realization, evaluating positions, playing time issues, and growth/decline expectations become an important consideration, if only to aid in the construction of one’s fantasy baseball team.

In the vein of considering organizational and positional talent, projection systems are a strong place to start. There is both a lot of unearned criticism of projection systems as well as over-reliance on them.  For a moment, a discussion on the manner in which projections should be considered. Projection systems, like Steamer which will be used later in this piece are integrated statistical models including numerous inputs like age, past production, and other factors. The projection is imperfect and is not meant to be relied on for exact truth but as the most probabilistically likely outcome among a diverse range of outcomes. In this sense, projections are a delightful benchmark, and because they lack the biases of those who watch a player or team constantly, they are important to challenge pre-existing assumptions. This is not to say that projection systems do not have a bias but rather that those biases or flaws are likely different than those exhibited by the eye test.

Therefore, surprising projections for Indians players from Fangraphs Steamer projection system for the Indians are detailed below. But first a final caveat, Steamer/600 is utilized which assumes a full season sample for each player. This is different than nob/600 projections which integrate playing time projections. I find playing time projections noisy, and ultimately assert that one wants to know the quality of play projected rather WAR totals in playing time projections which can fluctuate. Part one of this two part series will focus on the position player projections.

Below are projections for all Indians position players over 0.2 WAR in full time plate appearances. These are courtesy of Fangraphs and can be accessed here.

Courtesy of Fangraphs

*Ryan Ortiz and Austin Jackson are not currently in the organization.

As for what stands out there are two categories, risers and fallers.


Jose Ramirez: A more than 2.0 WAR decline from 2017 is projected in slightly fewer plate appearances. Where does the major decrease come from? Offense, specifically power. Steamer sees a 26 point decline in wRC+ which is massive, and a near 100 point dive in slugging percentage. Keep in mind, the slugging projection is still nearly 30 points higher than his career average which is where the pessimism comes from. 2017 was a massive power leap for Ramirez, and it is healthy to have some skepticism as to whether the full power gain will be retained or if Ramirez is merely a very good player instead of continual MVP candidate.

Michael Brantley: By using Steamer/600 this is at the outset an unbelievably optimistic projection for Brantley as the likelihood he reaches 600 plate appearances is probably below 20%. Further, while the offensive projection is solid, his defense is maligned by the system. Brantley’s defense has long been deemed poor by defense metrics and following yet another off-season of recovery from major surgery, this sort of defensive value decline makes sense.


Yandy Diaz: Steamer loves Yandy Diaz to the tune of a 110 wRC+ and a solid defensive projection. This is with Diaz posting a wRC+ of 86 in his only big league season, though with just 179 plate appearances. I have written previously on why Diaz projects as a player with roughly this sort of offensive projection:

Here is a line posted by a player in 2016 with batted ball components which are largely achievable for Yandy Diaz. Ground ball 55.6%, HR/FB 19% and Hard hit 35%. The home run per fly ball rate may be a bit high, but Diaz has far better contact% skills to compensate. Above, is SS/3B Jonathan Villar’s 3.0 WAR season in 2016. Of course, Diaz will be different, better in some ways and worse in others, but the profile is absolutely workable at the big league level to post lines around a .740 OPS and 100-110 wRC+, which is a boon since he can be deployed at multiple positions and used as a platoon killer.

Diaz has the athleticism, eye, flexiblity, and contact skills to make this profile work at the big league level.

Roberto Perez: Steamer relies on Perez’ elite defensive skills to carry much of this 2.2 WAR projection but also sees a substantial uptick offensively over the past two seasons. While Perez has the eye to carry a heightened offensive projection his contact skills would have to improve over the work of the past two seasons for this to become a reality. Considering Perez played much of 2016 injured, this may not be so far fetched. Still this projection is a testament to the depth of the catcher position for the Indians and the incredible quality of Perez’ receiving skills.

Once again, these are benchmarks to consider entering 2018 but each brings with it interesting questions for the players to answer over the course of the summer.

  • Steve

    As a FG writer, do you have any idea if Steamer or ZiPS will every be publishing percentile projections, similar to PECOTA?

  • tsm

    The obvious move, then, is to install Yandy at 3rd or in LF. Thus, moving Kipnis to 2nd makes no sense. We are stronger on D with Jose at 2nd. Kipnis could platoon in LF with Brantley or simply spell him in the event Brantley is slow to make a complete recovery. I suspect Yandy is more suited to 3rd than LF, but I will leave that to the analytics.

  • Tim Neal

    I like kip at second and jram at third I would teach Diaz to play first and patroon him there with Alonso and DH him when Encarnation plays first.

  • Mike Hattery

    Not sure, I am mostly at HBT at this point and do not know if they will take that approach. I certainly enjoy the revealing of PECOTA’s 25th percentile and 75th percentile projections.

  • Steve

    Considering that the team didn’t trade Kipnis or move him to LF and let Brantley walk suggests that their internal projections don’t see Diaz as the third best player on the roster. Some kind of combination of not having much prospect pedigree and his launch angle issues are legit concerns, unlike whatever the hell Francona has against his defense.

    Diaz has to be a pretty unique case, with the wide difference between the computer projections and the scouting ones. He didn’t make BA or BP’s top 10 before 2016 (not eligible this year), and was 17th for Sickels, who gave him a rating that equates to most likely a future role/bench player. By midseason 2017, BA moved him up to 8th (I don’t think the other two update midseason), but still behind not just Bradley and Chang, but the long-term projects that are Jones and Benson. Guys who are just about ready to become above-average major leaguers don’t get put behind good, but nowhere near elite, prospects in low A ball.

    I am one who trusts in the computer projections more than most here, but that so many other voices go against them does give us a reason to be skeptical in this specific case. There might have even been a chance to go Santana-Diaz instead of Alonso-Brantley, and they didn’t, and it’s not because they aren’t aware of how much better the computer projections are for the former pair.

  • Mike Hattery

    I think you can get Diaz a ton of playing time by merely platooning him with Kipnis/Alonso/Brantley, as well as resting Ramirez at 3B. Also, not super confident either Brantley/Kip exceed 120 games. A lot of time available.

  • Steve

    Thanks, and agree. Guys like Brantley, Kipnis, Alonso, and Zimmer would be much better explained with 25-75th projections, and not just the average.

  • Not a bad idea, but I worry that Yandy might have a tough time finding the required fifty settlers the patroon demands. That Dutch West India Company is notoriously unforgiving on the terms of a deal.

  • Emma

    Goo-g-le is payi-n-g 97$ per hour,wi-t-h weekly p-a-youts.You can also avail t-h-is.
    On tuesday I got a brand new Land Rover Range Rover from having earned $11752 t-h-is last four weeks..with-out any dou-b-t it’s the most-comfortable job I have ever done .. It so-u-nds unbelievable but you wont forgive yourself i-f- you don’t check it
    >>>>> http://GoogleHomeSuperCashJobsOpportunity/simple/work ✫★★✫★✫✫★✫✫✫★★✫✫✫★★✫✫★★✫★✫★★✫★✫★★★★✫★★★★✫★★★★✫✫✫✫✫✫✫★wx85!jwehfersd