it seems very odd to advocate applying a blanket team-specific Win adjustment (actual vs projected) which could punish/reward all the players on the team based on nothing but bad or good luck. for example, let's say a team is projected to win 100 games based off of their pythagorean record. but in each of the last three games of the season, the bullpen blows a 4-run lead and they end up only winning 97 games. so now, according to bill james, we should apply a -3% win contribution adjustment to each of the individual player's WAR because they "actually" only won 97 instead of 100?
I don't think he was doing that. I think he was positing the 6.8 WAR as being as potentially plausible as the 8.1 WAR.
― timellison, Saturday, 18 November 2017 23:33 (six years ago) link
Clutch hitting was a big part of why I thought Votto should have won the MVP this year
― timellison, Saturday, 18 November 2017 23:39 (six years ago) link
there are flukes in hockey, but not enough to win mvp. rob brown scoring 49 goals with an incredible 29% shooting percentage in 1989 comes to mind -- but he did that because he was on lemieux's line. and lemieux himself went 85-114-199. the great ones can give average/mediocre players a huge boost, but the latter will never overtake them
pitching and goaltending seem to be more prone to flukes. feel for the slider comes and goes, i guess
― mookieproof, Saturday, 18 November 2017 23:41 (six years ago) link
More evidence of the changing mindset of voters: this year's NL CY Young was only the second time a pitcher who led his league in both wins and ERA did not win (and the other time it went to Willie Hernandez during the we-love-closers era, so in a sense 2017 was the first time):
http://www.billjamesonline.com/2017_nl_cy_young_award/
Marks attributes the key to being IP; if Kershaw had won he would have also set a precedent, the lowest IP for a non-closer, non-strike-year winner ever. He also suggests H/9, but I don't know how much attention voters pay to that these days; I'd say the strikeouts were much more important (Marks mentions those too). There's a little bit on Scherzer and the HOF at the end, where he says pretty much what most of us have been saying here.
― clemenza, Sunday, 19 November 2017 15:13 (six years ago) link
Posnanski's response to James's piece:
http://joeposnanski.com/more-on-war/
(I believe they're friends, and Posnanski always writes favorably about James.)
― clemenza, Sunday, 19 November 2017 16:45 (six years ago) link
Doddering old guy makes the case for Charlie Blackmon (and, by implication, Arenado):
http://www.billjamesonline.com/mvp_followup/
― clemenza, Monday, 20 November 2017 12:27 (six years ago) link
cameron: https://www.fangraphs.com/blogs/putting-war-in-context-a-response-to-bill-james
― mookieproof, Monday, 20 November 2017 15:26 (six years ago) link
I hardly think the consideration of Pythagorean win-loss in James' article was being suggested as a "simple proposed fix." It was just illustrating a point.
― timellison, Monday, 20 November 2017 17:02 (six years ago) link
Players have flukey great seasons all the time, for example, because of a high BABIP that never levels out, even though it should, probabilistically speaking.
Are we also allowing for the possibility that a high BABIP over the course of a season is not always flukey, though, in the way that Joey Votto's high lifetime BABIP is presumably not flukey? (I know this post was from three days ago...)
― timellison, Monday, 20 November 2017 19:48 (six years ago) link