The forward-looking folks at Steamer Projections (@SteamerPro on twitter) have already released their 2016 forecasts, which are posted at www.Fangraphs.com.
They aren’t that promising for the Mariners.
For starters, Steamer projects Felix Hernandez to finish at 13-10 with a 3.22 ERA in 32 starts and a 4.3 WAR (wins above replacement) rating.
Nelson Cruz is ticketed for a major decline to 31 homers while batting just .255 (after hitting 44 and batting .302). Robinson Cano roughly projects to match his 2015 numbers with a .285/.344/.444 slash.
Premium content for only $0.99
For the most comprehensive local coverage, subscribe today.
Don’t be. At least not too much.
Because, well, listen (as former manager Lloyd McClendon loved to say), yes, these projections are the result of some really smart people crunching numbers (in ways, admittedly, beyond my comprehension).
And Steamer’s numbers are fun to examine and debate as the Mariners gear up for their first Hot Stove excursion with new general manager Jerry Dipoto. (Hey, I’m looking at them.)
The same holds true when Bill James publishes his projections in his soon-to-be-released Handbook 2016. (Aside: I’ve met Bill and believe he’s a genius who belongs in the Hall of Fame.)
All of that said…these are still projections, the most inexact of baseball sciences. (Quick now, a year ago, who had the Mets tabbed as World Series participants?)
Steamer’s projections are based on full seasons. They have James Paxton making 32 starts and going 12-13 with a 4.14 ERA. Projecting (roughly) a .500 record with an AL a bit over 4.00 isn’t unreasonable.
But you can believe the Mariners will bet Paxton exceeds those projections if he is somehow able to make 32 starts.
Yes, I know, these projections are based on every starter taking 32 turns and working 200 innings. And that doing so offers a standard base for comparison.
Similarly, Steamer has every reliever going 3-3 in 65 innings over 65 games. It also has Tom Wilhelmsen (3.84), Cody Martin (3.85), Rob Rasmussen (3.86) and Tyler Olson (3.87) compiling roughly the same ERA.
Anyone see that happening?