We are still just a few short weeks from pitchers and catchers reporting. Although the off-season has a been an unusual roller-coaster for the Queens ballclub, statistical modelers (including me) have been hard at work to get ready for the upcoming season. Projection models, refreshed for 2020, are starting to rapidly appear in the public domain – including the ATC projections.
I am the creator of the Average Total Cost (ATC) Projection system, which has appeared on FanGraphs since 2017. ATC is a “smart” aggregation model of other projection systems. The methodology behind ATC is similar to what Nate Silver does with his presidential forecasting at www.fivethirtyeight.com. Silver tracks polling data from various firms and assigns them credibility grades. He then intelligently combines the polling data; the better outfits receive more emphasis, and the inferior ones receive less weight.
For the ATC Baseball Projections, the same principal is employed and repeated for every statistical category. It is possible that for home runs – one system (System A) might get 20% of the total compliment, while another system (System B) might only account for 5%. Going to another category – say pitcher strikeouts – the total representations may flip. System B might earn 15% of the total, while System A may generate just 2% of the aggregate. And so on, and so forth for every single stat.
Others who attempt to aggregate projections – typically apply equal weight to all data sources. In terms of regression analysis, this is sub-optimal. Weights within ATC are based on the careful study of historical performance of its underlying projection systems. That’s what makes ATC “smart.” Also included is a blend of the past 3 years of historical data.
The Average Total Cost Projection system (ATC) has been one of the most accurate projection systems available to the public according to a FantasyPros study, as well as by a game theory analysis which I conducted last year on FanGraphs. I am happy today to share with you some early findings for Mets players.
Before getting to some of the 2020 ATC projections for the Mets, let’s look at some individual results from last year.
Player wRC+
Jeff McNeil 143
Pete Alonso 143
J.D. Davis 136
Dominic Smith 133
Michael Conforto 126
Brandon Nimmo 114
Todd Frazier 106
Wilson Ramos 105
Amed Rosario 100
Robinson Cano 93
Weighted runs created (wRC) is an improved version of Sabermetrics pioneer Bill James’ runs created (RC) stat. wRC attempts to put together all of a player’s offensive components into a single all-encompassing statistic.
The PLUS (+) in wRC+ indicates that the statistic has been additionally adjusted for park effects (some ballparks lead to more offense than others) and is also adjusted to league average. A wRC+ of 100 indicates a league average player. A 124 wRC+, for example, would indicate that the player had been 24% above league average.
With almost every single regular player on the ’19 Mets producing a wRC+ of over 100, the Mets were clearly an above average offensive team. The Mets scored a total of 791 runs last year, good for 13th in all of baseball. If the Mets can duplicate their 2019 offensive production, and if they can simultaneously improve upon their pitching (particularly with their bullpen), they will be in prime position to land a playoff berth.
Let’s now take a look at what might happen in 2020. As of today, here are the 2020 ATC Projections for
New York Mets hitters:
Player Name AB R RBI HR SB BA OBP SLG
Amed Rosario 574 70 64 14 19 0.275 0.316 0.418
Pete Alonso 565 96 108 43 1 0.256 0.351 0.541
Jeff McNeil 555 89 72 20 7 0.296 0.358 0.479
Michael Conforto 540 88 90 34 6 0.255 0.359 0.499
Robinson Cano 467 61 64 17 1 0.271 0.330 0.439
J.D. Davis 434 61 64 21 3 0.273 0.338 0.476
Brandon Nimmo 408 69 50 15 6 0.237 0.370 0.422
Wilson Ramos 407 47 62 15 1 0.279 0.337 0.437
Jed Lowrie 284 36 37 9 0 0.246 0.330 0.399
Dominic Smith 243 34 32 11 1 0.246 0.309 0.437
Jacob Marisnick 237 35 29 9 8 0.221 0.283 0.399
Yoenis Cespedes 196 29 32 11 3 0.259 0.324 0.491
Tomas Nido 163 15 18 4 0 0.222 0.266 0.343
Player Notes
Amed Rosario is projected to lead all Mets in at-bats with 574. Rosario is also expected to lead the ball club team in stolen bases, finishing with just under 20 – while hitting close to 15 round trippers.
There is no surprise that baseball’s HR champion and Rookie of the Year, Pete Alonso – is projected to lead of Mets in homeruns with 43. Second on the list is the veteran Michael Conforto (34) followed by second year Met and Statcast darling, J.D. Davis with 21.
Continuing with Alonso, ATC projects the slugging first baseman to knock in 108 runs, score 96 runs as well. Clearly, much of the Mets success in 2020 is dependent upon him.
Leading the team in projected batting average is none other than the flying squirrel and Zobrist-esque multi-positional player Jeff McNeil. In the baseball age of power and strikeouts, a .300 or near-.300 batting average is exceptional – and McNeil is projected for a .296 mark. After hitting .329 in ’18, followed by .318 in ‘19 – I believe that another .300 season is quite likely.
Speaking further of Jeff McNeil – after hitting 16 homers in the second half of last season, ATC projects him to hit 20 round trippers in 2020. If the newfound power stroke is a real skill acquired – McNeil may produce a near-MVP type season. The contact ability combined with his power surge reminds me greatly of a former Met – Daniel Murphy. Look to hear lots of comparisons this year of McNeil to Murphy. Jeff is truly essential to the Mets offensive campaign.
In his second season with the Mets, ATC currently projects veteran backstop Wilson Ramos to produce a statistically similar season. Last year, despite a 62% ground ball rate and hitting into 16 double plays, Ramos still managed to hit a lofty .288, including 14 HRs. Ramos has a career batting average of .275, which is elite for catchers. For 2020, ATC projects Ramos to compile a .279 BA (2nd highest on the team), with 15 home runs and 62 RBI.
Leading the team in projected on-base percentage is none other than Brandon Nimmo. The Mets would be wise to make him the team’s every day leadoff hitter [with McNeil batting 2nd]. In 2018, the following qualified players in major league baseball had a higher OBP than Nimmo – Mike Trout, Mookie Betts, and Joey Votto. That is the end of the list.
Finally, we get to Mets superstar Yoenis Cespedes. What can we expect from the former Citi Field HR Derby champion in 2020? The answer depends very greatly on the number of games that he will play. The ATC projections currently predict that Yoenis will get to roughly 200 at-bats in which he will hit 11 home runs, bat .259 and knock in 32 runs. That would be an excellent and productive pace for the injury plagued Cespedes, and we are hoping that the slugger can get to even more.
You can find the full 2020 ATC projections on the FanGraphs website under its projections section, or by clicking here.
