Quote:
Originally Posted by Chief Rum
(Post 2105872)
I like Edwin Jackson, but wins matter for this award (like it or not). 10 wins ain't gonna do it.
His teammate, Verlander, because of wins and that high K total, has a better chance than Jackson right now.
|
I'm not saying he should beat Greinke or Felix, but Jon Lester vis-a-vis Josh Beckett is in the same situation. Only up to 11 wins after his W in Tampa tonight, but had a FIP of 3.10 coming in (presumably going down) and is 3rd in K/9, 4th in K's overall now. When you consider he started 3-5 with a 6.07 ERA through May 30th, he's been as good as any AL pitcher the past 3 months yet only Beckett is mentioned as the Red Sox ace/Cy Young contender.
Quote:
Originally Posted by dawgfan
(Post 2105944)
On the specific examples you cite, I guess it all depends on how you judge the defense of those players; UZR would suggest that while Gathright is a plus OF, he's still well short of Gutierrez. Gutierrez has obviously taken a nice step forward offensively this year, but it could be argued he was already a better hitter than Gathright before this year. Anderson is worse than Gathright as a hitter and worse as a fielder, just a bit above average. Reddick, who knows - he doesn't have nearly enough major league time to measure his defense in comparison with the rest. It would appear that he has a shot at being OK with the bat though.
|
I don't think we're that far off at all. My (poorly worded) point here wasn't about any specific player (I picked the 3 on Pawtucket and you'll have to trust me that Reddick will be a good defender - above average range, elite arm) but that the replacement level for fielding is probably higher than the stats say - due to GM bias. BRAA works because we can assume that almost all hitters above league average are playing in MLB and a replacement hitter will not be very good. On the contrary there are multiple "proven" above average defenders bouncing around AAA because the perception they can't hit keeps them off a major league roster. Quite often the replacement when a starter goes down is actually an improvement defensively. It's not absolute, but starters are in MLB because they hit the ball better than the average player. Backups are there because they field better.
Quote:
You're right - Adam Everett doesn't compare to Kevin Youkilis. But who is saying so? FanGraphs isn't - they had Everett at a WAR of between 1.0 & 2.6 in his prime years (2003-2007), whereas Youkilis has been between 2.1 & 5.5 since 2006.
|
Yeah, that was kind of throwing a strawman out.
Quote:
I think the disconnect has been because defense is hard to quantify while offense isn't. Sabermetrics had offensive value figured out quite a while ago, with only incremental improvements over the last several years. Pitching on the other hand had the major bombshell of DIPS dropped in the last 10 years, and GM's are only starting to really fully embrace that discovery. Defensive metrics are still being refined, but while they're a long way from the accuracy of offensive stats, they've come a long way. There are still market inefficiencies with regard to defensive value, and there are some teams taking advantage of that. As defensive metrics continue to evolve though and you see more and more front offices that embrace sabermetrics, that inefficiency will close. The advent of Hit F/X will go a long way towards solidifying measurements of defense.
|
Hit F/X will help - and looking at the methodology for FRAA they take some of my earlier complaints like park effects and pitchers GB/FB into account - but it will only go so far because of the subjectivity involved. A hitter hits a ball and ends up with a concrete result - there are small grey areas between LD/FB for example if we're going for advanced metrics like BABIP, but pretty inconsequential ones - but it's not black and white for the defender, particularly in the OF. Even knowing the exact spot a hit falls, the speed it was hit at, time is was in the air, wind, possible wetness of grass, you still need to know the defender's positioning.
Quote:
Good stuff. One thing to consider though - Gutierrez is flanked by Ichiro in RF and for part of the season Endy Chavez in LF and now Michael Saunders. Ichiro and Chavez are both excellent OF's, and Saunders appears as though he is at least a plus defender. Balls that Gutierrez might otherwise chase down are being caught by Ichiro and Chavez/Saunders.
Ellsbury is surrounded by J.D. Drew and Jason Bay. Drew is a pretty good OF, but Bay is not. Ellsbury is getting some balls in LF that Gutierrez doesn't have to.
|
I don't think it really works quite that way. 1st the CF (should), like the SS, always have the right of way if they can get to a ball - it's rare I see a CF'er standing around or jogging near a FB being caught by another fielder, rather it's the other way around. And any slight added bonus of playing next to one bad fielder is probably negated by being forced to play out of position to compensate for them - or the opposite for a corner OF guarding the lines because of a great CF'er. (Supposedly UZR is broken down into either 6 or 12 OF zones alone - do they get broken out anywhere on the internets?) In the sense that positional alignments shift, outfield trio's are better judged together like as basketball units than as individual outcomes like batting is. Either way, arguments about specific players are getting off track from what's important. (EDIT - Haha - of course somehow Gutierrez had to go jump 3.4 RAR defensively in the interceding night while Ellsbury went even lower to 5th worst OF in MLB after Dunn, Hawpe, Bay and Wells.)
Quote:
Again, I think the crux of this whole discussion has to do with whether the kinds of ranges of defensive value put forth by the defensive metrics are realistic. Adam Everett in his prime was rated as being about 21 runs better than the average SS, and over the last couple of years Yuniesky Betancourt is rated about 14 runs below the average SS. You can argue with the specifics (has Betancourt been 19 runs worse than average this year?), but the more pertinent question is the general range.
And that's what I perceive as being the real issue here - it's easy to look at Adam Everett's batting stats and think there's no way his defense is good enough to make him an above average player overall, but if those metrics are correct, he is (or was). CF A could have an .850 OPS as a hitter but be bad enough in the field that CF B with a .700 OPS but a great glove could actually be contributing more towards his team winning than CF A.
|
I don't have a problem with the concept, and now that I look closer I think it was the WAR ratio to value that was throwing me off some. Using 2008 "qualified only" stats, they have 8 of 147 qualified players worth 30m+, 34 worth 20m+, 97 worth 10m+ and all of 6 worth a negative amount. Assuming mean distributions around 1/3 above each level, that's a minimum of 1.87b or 62+m per team - for less than 5 field players each. Adding in non-qualified players, there's another 28 worth 10m+ and we're out to 217 total worth 5m+. Pitchers add another 24 20m+, 75 10m+.
I looked and Dave Cameron explains this by saying that cost-controlled players are all underpaid and hence the marginal value per win should be 4.5m instead of 2.3m, but I'd argue that basic market economics implies that if they weren't cost-controlled they would drive down the overall market to equilibrium and teams all wouldn't suddenly double their spending. Really, this is semantics once you know what he's basing it on and now that I've looked (and mentally halved) every value it becomes more reasonable. I just wasn't seeing Morgan/Gutierrez - i.e. an elite defender with .750-.800 OPS - as worth 20m+, but 10m starts getting reasonable.