Every spring I collect several projections sources and average them together to do a meta-projection. Last year I pulled projections from Baseball Prospectus, ZiPS, Steamer, ESPN, and Baseball Reference.
When we wrote the projection article last year, Brandon Nimmo was just emerging as being saved in trade talks. For a chunk of time in January (before the Frazier signing), it looked like the Mets were going to deal Nimmo for to the Pirates for Harrison. When the season started, the Mets struggled to find Nimmo playing time in the outfield consistently (this was before the Cespedes injury and the Bruce at first experiment). Then Nimmo messed around and almost had an All-Star year:
2018 Average Projection: 427 PA, 250 AB, 70 H, 8 HR, .246/.339/.380
2018 Actual: 535 PA, 433 AB, 114 H, 17 HR, .263/.404/.484
(The difference between his PA and AB in the projections is due to some programs that previewed PAs seeing a lot more playing time than those which only project ABs)
Here’s what the projections got right: Nimmo was going to get on base a lot more times by walking than getting hits.
That’s where the being right stops. Nimmo hit for average nearly 20 points higher than projected, got on base over 50 points higher than projected and slugged more than 100 points than projected. Nimmo had a ridiculous year making him a key player for 2019 (and reason why Mets fans were complaining about possible Realmuto trades that involved him).
There was one program though that projected Nimmo out correctly for everything except playing time: Baseball Reference. They saw Nimmo hitting .264/.356/.421, his best slash line compared to all other programs. Next year they having him hitting .262/.380/.447.
