2/12/22

Paul Articulates – Finalizing the Mets’ 2022 Roster with an analytic approach

It looks like the baseball lockout is going to persist, as the MLBPA and the owners elevate their disagreements into the more public forum.  That means that there will be no official personnel action with MLB players such as trades, free agent signings, or salary arbitration.  During this “quiet period”, the team organizations are still very busy.  

The Mets hierarchy is working operational details, reviewing their farm system, player development process, and evaluating the international player market.  They also are working strategies for finalizing the 2022 rosters.    

In that vein, several recent articles in various publications state that the Mets are going to entertain offers to trade Jeff McNeil, JD Davis, and Dominic Smith among others.  Would you trade one or all of them?  Each has contributed greatly to this team in the past, yet each had a down year in 2021, fueling the speculation.   

In this post, I will take a "calculated" look at Mets position players that are not “established top stars”, but have shown the ability to contribute significantly to the team’s success.

For each position, there are several ways to evaluate a player’s contribution to the team and one metric does not work for all positions so it is prudent to use a blended approach.  Here’s some metrics that I’ll consider in the ratings of these players:

  •        “Off” – this is a metric used on FanGraphs to measure offensive runs above replacement that considers both contributions at the plate and contributions on the base paths.  I personally like this metric much more than WAR because of the importance of good base running in keeping a rally going.  Recent year performances are more important than prior years, so a weighted average of the last 4 full seasons is used.  Per FanGraphs, this is how MLB players stack up in the “Off” metric:

  • DRS – Defensive Runs Saved is a metric developed by The Fielding Bible that rates individual players as above or below average on defense by considering many aspects of a fielder’s range, ability, and effectiveness.  Recent year performances are more important than prior years, so a weighted average of the last 4 full seasons is used.  Once again per FanGraphs, here is the rule of thumb for DRS:
  • “Eye” – this is my own metric that looks at how much stress a batter can cause for the opposing pitcher.  In today’s game pitch counts and matchups dictate the replacement of a pitcher so if you can’t always get on base, you can at least stress the opponent with a savvy plan at the plate.  I use ratios of number of pitches seen to plate appearances and strikeout to walk to calculate “eye”.  Since this not an established Sabermetric, there is also not an established rule of thumb to compare with league averages.
  • Intangibles will also be considered, such as teamwork, position versatility, lineup versatility, fan support, leadership, and mental toughness.
Player value calculation: This calculation is customized to the Mets current make-up and to Citi Field.  In recent years, and demonstrated by the acquisition of Max Scherzer, this is a pitching-first team playing in a large, pitcher-friendly field.  Good defense is a multiplier for good pitching and makes every inning pitched by the starter count.  You need runs to win, but not as many as a pitching-weak team, so the defensive capabilities are weighted higher than offensive capabilities.  Intangibles are lightly weighted to be more of an “edge” than a dominant factor.  Position players on the 40-man roster without major league experience are excluded from this table.  

You may agree or disagree with the criteria, weighting, or individual evaluations, but at least it will give you a flavor for how Eppler’s staff may be figuring out how to proceed on the first day after the lockout ends.


Conclusions:



Even though we are not talking about the star players because they are not going to be put on the trading block, Francisco Lindor tops the value rating despite his sub-par year in 2021.  So let’s give him some respect – he will lead the Mets to much better things in 2022. 

Brandon Nimmo comes out of this analysis as a shining star.  His talents are under-appreciated amongst the sportswriters, possibly because they don’t recognize his continuous improvement in offensive and defensive areas.  He comes out on top of the outfielders for weighted DRS average over the last 4 years, has the second highest weighted Off, and is on or near the top of the WAR rankings on the Mets for the last 5 years.   I expect a lot from him in 2022. 

Jeff McNeil is clearly a “keeper” who is much more valuable in the New York outfield than he is on the trade block.  His offensive stats suffered last year during the confusion caused by the analytics staff briefings, but if you look at the “Off” rating in 2019 and 2018 he would be even more valuable. 

Luis Gillorme has an interesting outcome from this analysis.  I have always liked his infield versatility, but it also turns out that he has the best “eye” on the team.  An infield with Alonso, Lindor, Escobar, and Gillorme would make our aces sleep well at night.


JD Davis has a valuable bat, but his defensive contributions on a pitching-first team make him expendable – he should be on the trading block as the Mets seek another quality starter and a lights out lefty to replace Loup.  

Dom Smith is not going to help the Mets defensively outside of first base, and with Alonso justifiably holding down that position the Mets should seek to get trade value from Dom.  NY fans will miss him.  

Starling Marte, Mark Canha, and Eduardo Escobar are recent additions as the Mets’ front office flexed their financial muscles in the free agent market.   Escobar looks like a great add to the infield and Canha should factor in as a fourth outfielder providing better depth.  Marte has a surprising negative defensive trend over the past two years.  It is also worthy of note that this trio will be among the oldest non-pitchers on the roster and this could impact availability as the season wears on.

So would you do?




10 comments:

Hobie said...

Nice survey (given I still don't understand "OFF" stat.

Most surprise by the DRS stats of Pete, Nimmo ... and Nido (who too often is is just a default back-up rather that a real, albeit bench, defensive asset).

I've always scoffed at trade suggestion/projections as being as impossible as playing poker with yourself. It's just not an undertaking compatible with honesty. Given that's not what this essay was about (it was trade value) I'll yield to the temptation of thorough dishonesty.

McNeil, Dom, JD & Gillorme: ANY 4 to Oakland for Sean Manea & AJ Puk, ...OK all 4.
Then sign Bryant & Suzuki.

Tom Brennan said...

Hobie is my GM. Manea if healthy gets us a mighty long way.

Puk?? Not sure there - that has to be a gamble on potential.

Tom Brennan said...

Paul, I like to test out my model results against my guy - no way can Guillorme be # 4 and Starling Marte a # 7.

Luis has just 22 RBIs in 268 plate appearances. If he had lightning speed, I'd agree. But 3 career steals??

My gut has him ahead of only the 3 catchers. Hurry up, Francisco!

Tom Brennan said...

Paul, though, I like the analytical thinking, which I forgot to include in my last comment.

Paul Articulates said...

Tom, I was also surprised at Starling's result in the model. For the metrics I used, he has shown some decline over the last few years and the model is weighted to make recent performance more relevant than past. My gut agrees with yours, but I am going to take another dive in the next article about this spring's outfield competition.

TexasGusCC said...

Interesting exercise and certainly time consuming. We all have some stats that we try to navigate but there are things to consider here:
1. The players that weren’t on the team are affected by their previous team and cannot be held comparably to players that were Mets.
2. Smith is not an outfielder but pigeonholed into that position, and he did as good a job as a first baseman can do. There is no way Smith off of a great 2020 and a pretty good 2019 is the least valuable player, especially considering that Nido, Guillorme and even Mazeika are ahead of him.
3. As the DH is now a position, a DH’s fielding won’t matter. So, if Davis absolutely sucks with the glove, throw it away and let’s see how he does when that factor is discounted.

Tom Brennan said...

Gus, it will be interesting to see if a DH and a few other changes to freshen the team's blood will turn the collective offensive meltdown into a big rebound offensively in 2022. 2021 was not the hitters' year.

Tom Brennan said...

Paul, looking forward to it. Keep up the creative articles. I love to write creative stuff, even if no one else is, because being conventional is boring. You seem to have a knack for it, too.

Alex said...

It’s McNeil for Sonny Grey…..or McNeil for Urshela…

Remember1969 said...

This is an interesting piece of work, but there is something that just seems to be missing.

I am not sure how the 'Value Rank' is calculated, but it feels like any ranking system that has Alonso in the bottom half of the list behind Luis Guillorme and Jeff McNeil (not sure how Marte and Canha would really fall), and Travis Blankenhorn 3 slots above Dominic Smith is not quite complete. I am thinking that perhaps number of games played or plate appearances should be factored in some how.

Just the traditional straight up stats of BA and OBP, Blankenhorn was at .174 and .208 respectively in 2021 while Smith, in an off year was at .244 and .304. Blankenhorn had just 24 plate appearances. To use his -1 Off to Smith's -12.7 'off' seems to be incomplete. I just cannot see any way that he could be higher than Smith in any value ranking. Perhaps some sort of a normalization to some number of plate appearances would be valid.

Also, I don't think 2020 should be completely thrown away for this exercise. While I agree, it was only 60 games, those games still count to the players career stats. Both Smith and McCann showed their ceilings in 2020 and those stats should be included when looking at multi-year averages.

I think you are right on track when you talk about McNeil's offensive stats suffering at the hands of the analytics drive, but my suspicion is that the push to analytics was even more of a factor in the down seasons of McCann, Smith, and Conforto. (and yes, Lindor)

There did appear to be something different in the approach to hitting in 2021 across the entire team (I cannot explain how Nimmo was not as affected, while I think the system they tried to implement actually played into Alonso's already established strengths). I'm not sure averaging 2021 into a longer term sample size is valid for the Mets.

My last comment would be that I don't like to see common groupings or rankings include starting first basemen and utility infielders together. It just does not seem right to include Alonso and Guillorme, with their completely different roles, on the same list.