No Mirage in Arizona
7
|
The Hardball Times found this 8/20/2007 on www.hardballtimes.com [flag] |
Tags:
MLB
Arizona Diamondbacks
Comments (10)
-
notsellingjeans Great article, Chris. I think you've clearly established a second logical way that a team systematically outperforms its Pythag wins - terrible 6th and 7th relievers, and a manager who uses them correctly (thus making the blowout loss worse). The first way - "having a great 'pen" - is really too simple, and I've never read the point you espoused here before. That's brilliant.
I think you also hinted at a third way that a team systematically outperforms the Pythag, as well - the "feast or famine" starter, or perhaps a few of them. You know that my statistical chops don't match yours, but I look at Livan's game log, and I see an inordinate amount of blowouts for a guy with a 4.86 era. He's allowed 7-8 runs in a game FOUR times, and 6 runs on two other occasions.
In Livan's six worst starts, the DBacks have been outscored 62-15 - a 47-run difference! That's huge when you consider that, as you wrote, they've been outscored by 18 runs the entire year.
Interestingly, Livan hasn't allowed five runs in a game all season; every other start was 4 runs or less. Point being, he either pitches well enough for them to win or he gets absolutely rocked. He's the very definition of a feast or famine starter, and those guys will distort a team's Pythag. The paradox is, without Livan, they're a worse team, but might have a closer, more "predictive" run differential.
Let me postulate that the feast or famine pattern will express itself more with low-K, low walk, high contact guys - your Livans, Joe Blantons, and Carlos Silvas. They either face a Punch-and-Judy offense in a pitcher's park and pitch well enough to win, or they face a great offense/hitter's park and get crushed. There's not as much in between. They also don't bail themselves out of jams with strikeouts, or get yanked early because of bad control/high pitch counts. It's pretty much 5 innings and 8 runs with a two-homer inning that really got away, or 7 innings and 2 runs.
As a Bay Area guy for all of my 25 years, I saw plenty of these from Livan with the G-men, and now Blanton carries the torch. :) You're in bed by nine because you pretty much know the result by the fifth inning either way.
-
studes I don't mean to diminish your article, Chris, because it was nicely done. But I do want to mention to Jacob that this phenomenon has been pointed out many, many times before. The Yankees from two years ago, for instance.
I also very much agree with Chris that you're understating the impact of the Dbacks' record in close games. The Dbacks have the best record in one or two run games in the majors. They've had a great bullpen and tremendous clutch hitting. I have to do the math, but my guess is that this explains over half of their variance.
-
Chris Jaffe John,
I don't think 1-run games are just a coin flip. I think there's some correlation between it and team quality. I agree that the D-backs are exceeding what they should be doing in those games. Perhaps I understated it, but looking at their splits, the .371 mark in blow outs jumps out at me more than their .628 in 1-run games.
jeans - thanks fo the nice comments.
studes - I didn't know the core-periphery phenomenon had been pointed out so often in the past before. Shows how little I know.
-
bads85 The Diamondbacks have Bullpen WPA of 6.81, second only in the NL to the Padres 6.94. This appears to reinforce your assessment that, in general, the dregs are getting the ball when the game is out of hand, and the very good relievers are getting the ball when the game is on the line.-
studes That reinforces a pretty simple way to assess bullpen deployment: bullpen ERA vs. bullpen WPA.-
bads85 So let's take the Blue Jays, who have the second best bullpen ERA in the AL, but have a mediocre Bullpen WPA. Is their Bullpen ERA vs. Bullpen WPA suggesting the manager is screwing the pooch?
-
William Lu How about this idea:
Through Aug. 21 2007, Padres'and D'Backs' bullpen comitted impressive 6.94 and 6.81 WPA respectively. But the differential of BRAA is astonishingly more than 44 runs -- 57.97 VS 13.69 ...
Probably, it's safe to say D'Backs' bullpen are great in high leverage situation. Nevertheless, it's not necessary to say Bob Melvin made most of his decisions right but could be he really suffered more "clutch" situations than the other clubs, and forced Melvin to throw his best relieve arm in ..
-
studes The Jay's bullpen LI is actually higher than the Dbacks. I guess you could say the Blue Jays' manager is screwing the pooch, or you could say that it's B.J. Ryan's fault for not pitching up to expectations.
So I guess it's only partly bullpen deployment. The manager has to use his best pitchers in the best spots, but the best pitchers also have to execute.
Maybe it's good to not have a relief ace. Makes you more likely to be flexible and recognize your best pitchers.
-
-
-
-
GuyM Maybe a bi-polar bullpen is the key factor here. But just as plausible is the bi-polar starting staff. Or maybe something else entirely. Perhaps the poor record in blowouts reflects an offense that rarely scores more than 7 runs (in which case, it's the consistency of the DBacks' offense that really produces the overperformance).
To know if this theory is true, we need some questions answered. Like, is the D-Backs pen considerably more bi-polar than average? Does a bi-polar pen actually correlate with pythag overperformance in general? For now, all we have is speculation.
Links (7)
Baseball bonanza
Published 8/20/2007 by Beau Dure at Sports Scope
... (4.0 GB): Next: at N.Y. Mets, starting six-game East Coast swing. Reax: The Hardball Times says the D'backs surge is " no mirage , " citing a bunch of numbers dubiously attributed to Pythagoras ... NL WILD CARD 66-57 San Diego 65-58 Philadelphia (1.0 GB) 65-59 Atlanta (1.5 GB) 64-60 Los Angeles (2.5 GB) 63-60 Colorado (3.0 GB) 63-61 Milwaukee (3.5 GB)
Pickoff Moves
Published 8/22/2007 by Rob at 6-4-2 — an Angels/Dodgers double play blog
... Another factor in the decision:
[KWKW station president Jim] Kalmenson said another factor contributing to the deal with the Angels was the fact they have more Latin players on their roster than the Dodgers -- including star outfielder Vladimir Guerrero and pitcher Francisco Rodriguez -- and owner Arte Moreno is a fourth-generation Mexican American.
The Snakes Are For Real
Says Chris Jaffe:
Here’s how it works in Arizona. The game begins. If the starter pitches well, then it’s no problem. Wait until late, and let one of their dynamite relievers—Jose Velarde, Tony Pena, Brandon Lyon, and Doug Slaten all have ERAs under 3.00. And Juan Cruz is well above league average as well. Keep in mind they play in one of the game’s great hitters’ parks in a league that averages over 4.5 runs a game. As long as the starter has a quality outing, they’ll win a lot of g ...
Tuesday Links (28 Jul 07)
Published 8/28/2007 by Geoff Young at Ducksnorts
... No mirage in Arizona (Hardball Times, via ...
Game 141 Open Thread: September 10, 2007
Published 9/10/2007 by lboros <info@vivaelbirdos.com> at Viva El Birdos: Front Page Posts
... the entire series thumbed its nose at the concept of pythagorean records (and i'm a subscriber to the concept, by the way). arizona has been outscored by 29 runs this season and "should" be carrying a record of 69-75; they're 12 games better than that. the cardinals have been beaten by 62 runs this year and ought to be 12 games under, per pythagorus; they've topped their hypothetical win total by 10. . . . . a more outliery pair of september contenders you'll never find. a few weeks ago, chris jaffe of Hardball Times explained that the dbacks' ugly run differential is a mirage produced lagely by what he terms a "bipolar bullpen." when they're ahead or the games are close, the dbacks use their go-to relievers, who are all above league average for their respective roles; but when they fall behind, they go to a mop-up corps that is awful even by mop-up standards. hence the disparity between arizona's pythagore ...
rockies in 6
Published 10/8/2007 by lboros <info@vivaelbirdos.com> at Viva El Birdos: Front Page Posts
... the diamondbacks, it's well known, are a strongly anomalous team, having been outscored by their opponents 712-732 on the season. the Hardball Times' chris jaffe argues that the run differential is misleading, although his colleague sal baxamusa ...
The triumph of Pythagoras
Published 10/14/2007 by Pizza Cutter at Statistically Speaking
... So, are the Arizona Diamondbacks a sub .500 team, like their Pythagorean projection says or are they a 90 win team like their… ummm… actual record says? It’s an interesting question. When trying to figure out how “good” a team is, which should we look at? This is a topic which has been taken up before by Chris Jaffe, specifically with reference to the Diamondbacks, and ...
Tracy, Tracy, Tracy…
Published 10/16/2007 by Bill Baer at crashburnalley.com
... posite direction. 19 of the 25 players on the post-season roster of the Boston Red Sox are not home-grown. Everyone: don t home-grow your talent! And they [won 21 of 22 games; the Wild Card; the NLDS; the NLCS] as a team. As opposed to those un-home-grown Red Sox, who won the AL East as individuals. When the Red Sox were jumping up and down on the mound after clinching the East, you could hear Jason Varitek screaming, Yay me! and see Jonathan Papelbon holding a giant foam hand with I m #1 written on it. While Arizona manager Bob Melvin sits back and watches like he s in spring training, making sure people get their work in[ ] While Melvin s D-Backs are indeed a fluke (check out their run differential), this snipe at Melvin is unwarranted. I m sure Tracy knows this, but the D-Backs won their division with a 90-72 record, best in the National League. I m pretty sure Melvin doesn t approach any game like a spring training game, let alone one in the post-season. And, if anything, ...
Chris
I really enjoyed your article on the D-backs. I have a couple of thoughts:
1) I'm not sure I agree on the 1-run game analysis. If you assume that 1-run games are a coin toss and omit them then Arizona's record is 44-38, which explains 40% of their deviation from Pythag - in my boook that is quite a lot.
2) The other reason that teams overperform their pythag is because of the bullpen so it is no surprise that the Snakes have a great bullpen, as you point out. The split between the core and the dregs is interesting but I don't think is revolutionary. Ideally you don't want any dregs in your bullpen whatsoever but if you have them then use them in low leverage situations. Now apart from the closer debate I think that this is, save a couple of egregious examples, standard practice.