Discussion in 'Entertainment, Culture, and Sports' started by edinatlanta, Feb 3, 2011.
Send me your address. Or are you still all over the place?
LEYLAND IS COMING BACK.
And his staff and his sisters and his cousins and his aunts. (From "HMS Pinafore")
I would be happy if they announce that Brookens will be the new 3rd base coach.
Dotel and Perralta have been signed.
So here are the GG winners they have on MLB.com
C: Matt Wieters, BAL (2nd award)
1B: Mark Teixeira, NYY (5)
2B: Robinson Cano, NYY (2)
3B: Adrian Beltre, TEX (4)
SS: J.J. Hardy, BAL (1)
LF: Alex Gordon, KC (2)
CF: Adam Jones, BAL (2)
RF: Josh Reddick, OAK (1)
P: Tie: Jeremy Hellickson, TB (1); Jake Peavy, CWS (1)
C: Yadier Molina, STL (5)
1B: Adam LaRoche, WAS (1)
2B: Darwin Barney, CHC (1)
3B: Chase Headley, SD (1)
SS: Jimmy Rollins, PHI (4)
LF: Carlos Gonzalez, COL (2)
CF: Andrew McCutchen, PIT (1)
RF: Jason Heyward, ATL (1)
P: Mark Buehrle, MIA (4)
And here's how Defensive Runs Saved (my personal fave of the fielding metrics. Straight forward, tied to reality, easy to explain) ranks the positions:
AL, P: Ricky Romero (6) (Peavy at 4, Hellickson at 2)
NL, P: Mark Buehrle (12) and Jake Westbrook (11)
AL, C: Alex Avila (6) (Wieters was just behind at 5)
NL, C: Yaddy (16)
AL, 1B: Tex (17)*
NL, 1B: Votto (9)* (LaRoche just behind at 8)
AL, 2B: Cano (15)
NL, 2B: Darwin Barney (28)
AL, 3B: Brett Lawrie (20) (Beltre was 3rd in AL, 4th overall at 13)
NL, 3B: Wright (16) (Headley was -3)
AL, SS: Brendan Ryan (27) (Hardy is #2, but pretty far back at 18)
NL, SS: Clint Barmes (13) (Jimmy Rollins was 3rd worst overall, 2nd worst in NL at -8)
AL, LF: Alex Gordon (24)
NL, LF: Martin Prado (12) (CarGo was dead last in baseball among qualifiers at -13, tied with Josh Willingham)
AL, CF: Denard Span (20) (Jones was -16, worst in baseball)
NL, CF: Michael Bourn (24) (McCutchen was -5)
AL, RF: Reddick (22)
NL, RF: Heyward (20)
*NOTE: A-Gone, who split time between leagues, was at 16.
So the usual mixed bag for GG results. 8/18 matched with the #1 player by DRS. Call it 11/18 if you give them AL-P, AL-C, and NL-1B which were very close. Another 2 (AL-3B and AL-SS) the margins were wider, but they still picked a top-flight defender, so that's 13/18 pretty darn good ones.
But then there were some atrocious ones. Headley was 19 runs worse than Wright, Rollins was 21 worse then Barmes, CarGo was 25 worse than Prado, Jones was 36 worse than Span, and McCutchen was 29 worse than Bourne. And just to be clear, this wasn't just a particular defensive stat not liking them. UZR still shows them as bad defenders (comparatively in some cases, just overall in others):
Headley: 9 worse than Wright (who's still #1), 6 to 15.4
Rollins: 10 worse than Barmes (who's still #1), 4.4 to 14.4
CarGo: 19.2 worse than Prado (who is narrowly #2 in UZR to Soriano and CarGo is still the worst in baseball) -8.5 to 10.7.
Jones: 15.2 worse than Span (Span still #1, Jones now only 6th worst in baseball, 2nd in AL) -6.7 to 8.5.
McCutchen: 29.3 worse than Bourn (Bourn still #1, McCutchen actually looks worse) -6.9 to 22.4
Rollins and Headley don't look as bad as they're at least positive under UZR, but still clearly behind guys who are #1 under both metrics. And then CarGo, Jones, and McCutchen are just atrociously bad GG winners.
Headley: 19 worse by DRS, 9 worse by UZR (14 average)
Rollins: 21 worse by DRS, 10 worse by UZR (15.5 average)
CarGo: 25 worse by DRS, 19.5 worse by UZR (22.25 average)
Jones: 36 worse by DRS, 15.2 worse by UZR (25.6 average)
McCutchen: 29 worse by DRS, 29.3 worse by UZR (29.15 average)
Michael Bourn was arguably the single most valuable defensive player in the league last year. He was #3 across ALL POSITIONS in Defensive Runs Saved at 24 and was #2 in UZR at 22.4 (only losing out to teammate Jason Heyward by .5 runs). Think about that for a second. Two separate, well-crafted metrics agree pretty tightly that Bourn saved his team nearly 30 runs with his glove more than McCutchen. That's roughly 3 entire wins worth of value. And he lost a fielding award to a guy (in McCutchen) whom both metrics agree wasn't just worse but was actually even losing his team runs in the field.
Fuck me I really hate sabermetrics.
Trout was not included in the lists I looked at given his limited playing time, but he's actually a bit better than even Span who I listed at #1 (23 DRS in CF, 10.6 UZR compared to 20 and 8.5).
Why? I talked to two separate Orioles fans on another forum, for example, and they both agree the Jones pick for GG makes little sense.
Because it sucks the life out of baseball. Though I will also agree that Jones didn't deserve it with the caveat that I'm a Yankees fan.
How so? I've heard this said many a time but never explained. People have used stats to explain baseball since it was created. How does using better (aka measured/tested/evaluated/tuned) statistics suck the life out of anything?
The 'eye' test for fielding is one of the best examples of how sabremetrics helps. Analyzing fielding with your eye and nothing else is nigh impossible. It's game speed, there's almost certainly selection bias as even die-hard fans rarely see 100% of a player's fielding time, huge memory biases where you remember Web Gemmy spectacular plays even though they add no additional value compared to making said play normally, and most importantly you're trying to measure a negative (plays not made) with nothing to compare it to.
Everybody knows that the "traditional" stats are flawed and don't show the whole picture. Sabermetricians claim to, or at least strive to, capture the whole picture in numbers. Cold, dead numbers.
because sabremetrics isn't only used to analyze past performances, it's used to predict future performances. according to sabremetrics, the orioles should've been a losing team.
Every team other than the Giants is a losing team, my friend.
Right... How does that suck anything out of the game? Everybody uses past performance to predict future performance. Again, don't see how doing so differently and likely more accurately destroys the game or whatever.
By the by, run differential isn't really sabermetric, it's pretty basic. If you score fewer runs than you allow you're going to be a losing team in general.
Also no, the O's scored 7 more runs by season's end than they allowed. That's essentially a .500 team. A basic Pythagorean win/loss record pegs them at 82-80, and that generally has a 4ish game MoE. Performing 11 games above said Pythag record is crazy, no other team in baseball came close to that. But the Orioles were, overall, not a losing team. They put it together in September (getting healthy) to make a strong final run. The thing that annoys some folks is that they were able to capitalize on that strong run so much because of the benevolent position they'd been left in to that point because of, yes, some serious luck.
After August 31st the O's had scored 554 runs and allowed 593, but were 73-58 (.557). That's absurd. And yes, using past performance to predict future performance, it's not silly to say that a team getting outscored by 40 runs (nearly a third of a run a game) would not keep getting such strong results. But the underlying team changed, so their expectations did.
Literally no actual sabermetrician claims that. Have you read ANYTHING on the subject by people actually involved in it? They're basically universally careful to admit the flaws in their measures, the margins of error, and the things stats just can't pick up. That's the beauty of actually discussing how you see the game, it allows you to notice flaws in your own analytics. Your post is the worst kind of strawman argument.
You need to calm down there Billy Beane. I feel like you've posted sabermetrics before and got similar responses. Great, go play with Stata to your heart's content but don't take offense when people don't like it.
The fact that Trout got shafted bodes ill for his MVP chances.
Separate names with a comma.