By Andy Singer
December 14, 2022
As often happens here at SSTN, a lively conversation started the other day in the comments section comparing the Yankees to the Mets as the two rosters stand today. Without looking at any numbers, I opined that the Mets might not be better than the Yankees as they stand today, even with all of the flashy spending the Mets are doing under Steve Cohen. Our Editor-In-Chief, Paul Semendinger, had a very different perspective, noting that the Mets were winning not just the PR battle (a point I'll grant without argument), but the roster building battle as well.
To illustrate his point, Paul compared the two teams by looking at them position by position using a mix of bWAR and traditional stats from the 2022 season. I recommend reading it, as I think it was fun to look at two teams in that way, as I would comparing two teams at a bar with other fans. I am not going to say that Paul's methodology is invalid, but I don't think it is the most precise methodology that can be used to project two teams for the future. My perspective is that looking just at stats from 2022 doesn't give you a full or predictive picture of either team. I also don't think that going position by position paints the most instructive picture, because it doesn't account for the total value produced (or detracted) by an individual player. WAR totals, while not perfect and not the be-all-end-all, is the best method we have to calculating total value, so I will lean heavily on that statistic.
I wanted to look at what one of the major projection systems had to say about the two teams. The best publicly available projection system that has its projections for the 2023 season posted is Steamer. Steamer projects playing time based on the roster around a player and takes into account the relative "maturity" of a prospect. Most importantly for our discussion, Steamer's projections correct for recency bias by taking a player's total body of work into account. For full disclosure, I don't think Steamer projections are perfect, and I do have the following critiques:
Steamer does not weight injury history heavily enough when projecting playing time.
Bullpen projections are very depressed, meaning that total bullpen values are very jumbled (so you won't see any bullpen charts, just overall impressions).
It takes a very large body of work for starting pitchers to break through in the projections.
With that out of the way, the Steamer projections for 2023 paint a fascinating picture! I won't spill all the beans just yet, but Steamer shows that the two teams are remarkably close as it stands today.
As a whole, Yankee position players project to produce 33.1 fWAR whereas the Mets project to 30.9 fWAR. I did not expect to see that, but there it is, and it makes sense after digging a bit deeper:
Yankee position players are helped tremendously by Aaron Judge and his 6.9 fWAR projection for 2023 (2nd best in Steamer's projections, just 0.1 fWAR behind Juan Soto). There is little question but that the Yankees as currently constructed are very dependent on Aaron Judge to remain competitive, but he has remained healthy the last couple of seasons, putting some fluky injuries in the rear view mirror.
The Mets have their 4-5 top players, and then it's a very steep drop-off, with almost no depth in the event of injury.
That may sound familiar to Yankee fans, but the reality is that the projections show a glut of players who can project to roughly average on a per plate appearance basis in total value. Even players like Josh Donaldson, who is a poor value for his contract, is a good bet to produce at an average level when defense is considered.
Behind Aaron Judge, the Yankees have a couple of very good players. Many of us are hot to trade Gleyber Torres, but he is the second best position player in the projections for the Yankees (4.1 bWAR last season and Steamer projects 3.5 fWAR here). Likewise, Harrison Bader projects to 2.9 fWAR with all-world defense and slightly above average offense, which would actually be a step back from his actual performance in 2020 and 2021, his last fully healthy seasons.
Giancarlo Stanton is projected to bounce-back, with numbers right in line with his career performance (132 wRC+). The truth is that Stanton was hitting to those numbers last season until 134 bad, injury-induced plate appearances tanked his stats. Our own recency bias overweights recent statistics in this case.
All of us wanted Francisco Lindor, but the combination of 1/2 seasons from Anthony Volpe and Oswald Peraza are projected to be worth 3.2 fWAR (!). That's pretty lofty for rookies, but Steamer shows that the hype is statistically warranted.
The Mets are super dependent on a healthy Brandon Nimmo, their best player, given that they don't have a lot of depth. The projections credit him with 689 plate appearances, but he hasn't come close to that in any season other than 2022. Nimmo's last season with even 500+ plate appearances? 2018. That should really scare Mets fans, as every other season is below 400 plate appearances.
The Mets don't approach average at 3B, C, and DH. The Yankees have pieces that should obviously be upgraded, but they at least have plug-and-play options who would be acceptable in the regular season statistically. At DH, the Mets had a late season hero in Daniel Vogelbach who doesn't project well as a platoon hitter that can't hit lefties and doesn't have a platoon partner as of this post.
Mets starting pitchers project to 15.2 fWAR whereas the Yankees project to 13.3 fWAR, closer than I think most fans would predict, given the Mets' flashy moves. A couple of notes:
This projection does not include Kodai Senga, but who knows who Senga will be. Scouts across the league had significant questions about Senga's ability to transition to MLB from NPB due to his relative lack of command, lack of strike throwing in NPB's more pitcher-friendly strike zone, lack of a third pitch, and the potential of his forkball to translate to the MLB ball. For this reason, I don't feel comfortable projecting him either.
After their top two pitchers, one of whom is in his late 30s and has begun to miss time due to injuries, while the other is 40 years old, the Mets have a steep drop-off again. Quintana just experienced a dead cat bounce in 2022, but hasn't performed to that level in any recent season since his days with the White Sox, and Carlos Carrasco has a long injury history and is 35. The Mets are showing a lot of faith in the health of a very old pitching staff.
The top end isn't as stellar as the Mets, but the Yankees' projected rotation is less risky and volatile. Gerrit Cole still projects to be a soft ace, and the collection of pitchers behind him project to be above-average, and it's easy to squint and see getting more out of Nestor Cortes and Luis Severino. Likewise, we have to see if Montas can pitch in pinstripes, but the skill level is there for that to happen.
Mets relievers as a whole project to 3.3 fWAR to the Yankees' 2.7 fWAR. Generally, the projections show that the Yankees have quantity and quality in the bullpen, though without an arm that projects to be in the top-15 relievers in the sport (again, Steamer really depresses the value of relievers). The Mets have an elite closer and largely an island of misfit toys behind him, so it is a huge risk if Edwin Diaz falls off a cliff (which, by the way, has happened before). I'd take the Yankee bullpen despite the Steamer projections.
The Mets project to 49.4 fWAR today. The Yankees project to 49.1 fWAR. I think that functionally, the two teams are pretty close to even, but the Yankees are clearly not done yet. With a couple of moves, I have no issue saying the Yankees are better than the Mets. I also think the range of outcomes for the Yankees are less volatile than the Mets due to depth and an over-reliance on a few stars.
I think both teams will be excellent, to be fair, but there is clearly still work to do.