A Look at One Writer’s American League Rookie of the Year Ballot

I had the honor of voting for this year’s American League Rookie of the Year award, and the biggest challenge was — not unpredictably — how to weigh performances over a 60-game season. Adding a layer of difficulty was the fact that some of the best numbers were put up by players who weren’t with their team for the duration of the campaign.

Willi Castro and Ryan Mountcastle excelled with the bat — especially Castro — but each had only 140 plate appearances. Sean Murphy, who augmented his solid offense with strong defense behind the plate, had exactly that same number. Are 140 plate appearances enough in a truncated campaign? Following a fair bit of deliberation, I decided that they aren’t. As a result, all three players fell off my consideration list.

And then there were the pitchers. Not a single rookie in the junior circuit threw as many as 65 innings, and the most dominant of the bunch totaled just 27 frames. This made for an especially difficult dilemma. Would it be reasonable to give one of my three votes to a lights-out pitcher whose relative workload was akin to that of the position players I’d chosen to discount? Moreover, had any of the higher-innings hurlers done enough to preclude me from making what amounts to a contradictory choice? We’ll get to that in a moment.

Given that Kyle Lewis garnered all 30 first-place votes, and that Luis Robert garnered nearly all of the 30 second-place votes, it’s obvious what the top of my ballot looked like. The ordering wasn’t a particularly hard decision. While one could reasonably argue that Robert’s defense was as valuable as Lewis’s offense — their respective WAR totals (Lewis 1.7, Robert 1.5) reflect that — the Mariners’ outfielder dwarfed his White Sox counterpart in both wOBA and wRC+. Like my BBWAA brethren, I opted for the better offensive production.

That brings us to pitchers, my third-place vote, and the most time-consuming part of the process.

Whittling down my initial pitcher list didn’t take terribly long. Among those considered, but ultimately lopped off, were Pete Fairbanks, Matt Foster, Codi Heuer, and Cristian Javier. That left two pitchers to choose between, and to say I dithered on that decision would be an understatement. Truth be told, I probably changed my mind half a dozen times.

Justus Sheffield threw the third-most innings among AL rookie pitchers, and he had a 3.17 FIP to go with a 3.58 ERA. (By comparison, Javier had a 4.94 FIP and a 3.48 ERA.) Moreover, the Seattle southpaw’s 1.6 WAR was tops among his rookie-eligible pitching peers.

James Karinchak tied for the most appearances (27) among AL rookie pitchers, and his 1.1 WAR was bested only by Sheffield’s total. Most impressively, the Cleveland reliever allowed just 14 hits and fanned an eye-popping 53 batters in 27 innings.

I solicited opinions before deciding between the two. I checked in with several FanGraphs colleagues, as well as a handful of highly-respected writers from around the country. The dozen or so recommendations varied, albeit with a clear consensus: Karinchak merited my third-place vote. And when push came to shove, that’s whose name I opted to check. Small sample size or not, Karinchak’s dominance was too hard to ignore.

My ballot: 1. Kyle Lewis, 2. Luis Robert, 3. James Karinchak.





David Laurila grew up in Michigan's Upper Peninsula and now writes about baseball from his home in Cambridge, Mass. He authored the Prospectus Q&A series at Baseball Prospectus from December 2006-May 2011 before being claimed off waivers by FanGraphs. He can be followed on Twitter @DavidLaurilaQA.

newest oldest most voted
sadtrombone
Member
Member
sadtrombone

I guess I don’t understand the “they didn’t play enough” argument. If someone only gets 140 PAs and puts up as many wins as the guy who got 240 PAs, then they were approximately equally valuable, no? When you look at counting stats it inherently takes into account playing time, so it seems like you’re double-correcting for it.

I’m not thinking that Castro or Mountcastle deserve to be on the ballot anyway, but Murphy was awfully good and nothing he did looked particularly fluky.

MikeS
Member
Member
MikeS

Arguably, they were more valuable since someone took those other 100 PA and were probably above replacement value. It’s sorta the same argument for why you are better off with one 4 win player instead of two 2 win players.

D-Wiz
Member
Member
D-Wiz

I don’t think that’s the right way to look at it… If a player is more valuable the less they play, then shouldn’t they not play at all? I think playing time matters – it’s trite, but “the best ability is availability” is true in some sense. The knock-on effect means that the roster is stronger overall when your best players play the most innings. Now, I do think that Sheffield should have gotten the 3rd place vote over Karinchak for that reason – Sheffiled threw twice as many innings, which means there were twice as many innings his team didn’t have to use a worse pitcher.

Jim
Member
Member
Jim

Any chance you are the DD Wiz that comments on GoComics? If so, they are excellent.

D-Wiz
Member
Member
D-Wiz

As much as I want to take credit for anything “excellent”, no that is not me lol.

Jim
Member
Member
Jim

Thanks. I still like your comments here, of course.

The Ancient Mariner
Member
The Ancient Mariner

Actually, no, you can’t assume that whoever took those other plate appearances was above replacement value. That doesn’t actually hold–probably for the same reasons that the best evidence indicates teams *aren’t* actually “better off with one 4 win player instead of two 2 win players.” (If they were, the idea of $/win would be a complete non-starter.)

Cave Dameron
Member
Cave Dameron

Isn’t one of the main purposes of WAR is that it aims to eliminate any bias held towards traditional counting stats and games played/innings/at-bats. The same value spread out over more games isn’t really more valuable is it?

bosoxforlife
Member
Member
bosoxforlife

One of the things that I have noted since becoming involved in chatting on the various platforms is that analytically biased people are more likely to actually hold exceptional performance in the traditional counting stats against a player and rather strongly. Jose Abreu is clearly treated this way.

sadtrombone
Member
Member
sadtrombone

My point is that WAR is a counting stat, although not a traditional one. It’s not a rate stat. If you put up 1.5 wins over 30 games or 1.5 wins over 60 games, you still put up 3 wins. So why penalize someone for not playing 60 games? That’s already accounted for.

I think it’s also worth talking about your second point: “The same value spread out over more games isn’t really more valuable is it?” In terms of roster construction, it’s more valuable to have concentrated WAR in one position (so having a 5-win guy at third base versus 2.5 win guys at 2B and 3B), but in terms of award voting I think I agree with you. It is probably a wash to say it was better/worse because it was concentrated in fewer games. I’m sure someone here could model but it would be a waste of time because the conclusion would be “it depends!”

Joe Joe
Member
Member
Joe Joe

From 2012 through 2020, the Angels have won 682 games and WAR suggests they should have won 684. As Mike Trout has been the most concentrated source of WAR over that time period, one would expect to see some signal if concentrated WAR helped win games.

When you start talking guys closer 0, the error in assumption of replacement level and roster construction issues matter, but 2.5 WAR is such that almost every team in history could find room for two 2.5-win players. Maybe the Dodgers have a hard time getting playing time for two 2.5-win players, but they aren’t the entire market.

MikeS
Member
Member
MikeS

I think that just means they have been utterly awful at placing pieces around Mike Trout, despite lots of opportunity for improvement.

sadtrombone
Member
Member
sadtrombone

Yeah, a better way of putting it is that concentrating WAR in a single player is better because at some point you run out of positions to put 2-win players. If you have a bunch of 1.5-2 win players, getting two 2.5 win players is a modest upgrade but a 5 win player is a big one.

This presumes you have a bunch of 1.5-2 win players. Cleveland and the Angels have so many 0-win positions that they’ve utterly failed at taking advantage of it.

Joe Joe
Member
Member
Joe Joe

I get it if you want to talk 1-win players, but 2.5 win…come on. The assumption of having a bunch of 1.5-2 win players does not encompass many teams. Generally, the teams (not counting Dodgers/Yankees) that have a bunch of 1.5-2 win players get there by going by trading concentrated WAR for less concentrated WAR and more years of it/having draft strategy targeting floor over upside. The Angels continue going for concentrated WAR and their minors is loaded with flashy position players that have trouble making contact.

sadtrombone
Member
Member
sadtrombone

I mean, it’s not that complicated. If you have two 1.5 win players at second and third getting 600 PAs, and you get two 2.5 win players to replace both, you’ve added two wins. If you get one 5 win player to replace one, you’ve added 3.5 wins. This is why 2-win guys like Kyle Gibson get $28M and 6 or 7 win guys like Gerrit Cole and Mookie Betts get $200-350M. The 1.5-2 win guys don’t grow on trees but they also aren’t that hard to find.

baubo
Member
baubo

The reason why a 4war player should in theory be more valuable than 2 2war players, or in this instance a 2war player in 30 games be worth more than 2war player on 60 games, is that good teams are capable of finding better than replacement level players when their starter goes down.

However, this assumption doesn’t apply to badly run teams where they truly run out players that are simply bad. The Angels have had the ideal situation where they could get 8 concentrated war from Troutx but they couldn’t use it since they cant get those cheap pitching value the way the A’s and the Astros have. Put Trout on the Astros or the A’s in place of Laureano or Springer and one can see just how valuable he truly has been

Joe Joe
Member
Member
Joe Joe

Correct. Outside of the low extreme, fWAR usually does a great job of calculating value such that more WAR helps a team win more regardless of PA or innings.