managers' effects on player performances using this type of metric. Great call :-)
Our old friend Steve Nelson used to do a great job reminding us that --- > we get too tempted to figure "18 guys will have normal years and 7 guys will have good years." In reality, said Steve, NOBODY is going to have the same year as last year. Figure on as many of them going down, as going up. Unless your team is very young.
Whattaya know, this last week Bill James did some studies on that question. He's got two studies up as to the questions, "When a team wins the pennant, how many of their guys have good years? How many NEED to have good years?" and related. He set up the study so that the "good years" and "bad years" balanced out 50-50 for the whole league. Felix last year had a "bad year" versus his own career standard.
Oversimplifying the answers, and bearing in mind that James might not agree with my interpretations of his own work:
1) You don't NEED to have a 60-40 split of good years vs bad, if you already have the best team.
2) "Cinderella" teams, legendary teams that did far better than expected, had a good 67-33 split. Two-thirds of their players surprised in the UP direction, one-third in the DWN direction. The 1950 Philly Whiz Kids had +11 players, -6 players. The 1969 Amazin' Mets had only +12 up, -11 down, though. The 1984 Tigers -- didn't they start 30 and 5, or something? -- had +15 -6. The 116-win Gillick Mariners were an astonishing +20 -5. Just for ol' Rockies Jeff Gwinn ... the 2007 Rockies were +19 -8.
3) Very disappointing teams tended to be even with their hitters, but way negative on their pitchers. A delightful little surprise of sabermetrics, if it holds up. The 1988 Orioles - were they they team that ripped off 21 losses to start the year? Had one pitcher with an UP year, and 9 pitchers with bad seasons by their own career standards.
4) The rule saying "You win when your players have good years together" is nowhere near absolute. The 1952 Browns lost 90 games with 9/10 of their hitters having good years and 3/6 of their pitchers having good years. LOL.
Last year's Cubbies were +17 -6, not that they'd have NEEDED that. They won their division by +17 games because of the intersection of the best talent and a bunch of good years.
By the way, this paradigm caused James to suddenly realize that you could measure managers this way, and he opined that Terry Francona would do great. That managers cause players to have UP years. Dr. D crushed his head against the desk. In two sentences, twenty years' worth of the "Chemistry" debate inverted itself.
As it should.
... further ...
Ol' Jeff??? Okay....old is right. Can't argue with it anymore!!! The key to the 2007 Rockies were that I bet the last three weeks of the season they were 25-0. Matt Holliday was worth about half of that. They all got hot at the right time....until they ran into the Buzzsaw in Boston.
2017 Mariners? I think some guys will be hard pressed to outdo last year. But there is a lot of room for some improving numbers, especially in the bullpen.