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.
Todd Frazier was billed as the key signing for the Mets last year. Their one signing that wasn’t a member of the previous Mets team, a signing that was going to push Cabrera to 2nd and save us from Reyes or Cabrera at third base.
2018 Projections: 562 PA, 494 AB, 109 H, 28 HR, .232/.321
2018 Actual: 472 PA, 408 AB, 87 H, 18 HR, .213/.303
No individual projection (you can see them here) was able to accurately predict the fall-off on Frazier’s numbers off his career numbers, let alone his projections. In fact, last year felt so abnormal that if Frazier was going to have normal playing time next year, he’d probably be a pre-season candidate for comeback player of the year, contingent of course on this slide being a fluke and not a trend.
Frazier this upcoming season is being pushed into a utility role with the Mets bringing in Cano and Lowrie who should be playing every day. It will be curious how computer projections try to square this for Frazier in 2019.
