The thick lickquid Gino Marchetti searched the world over for…Goldman serves up here.
My pals at Baseball Prospectus have a little congeries of stats we call the Secret Sauce. Introduced in the book, Baseball Between the Numbers, to which your humble host contributed a chapter and a little page about the relationship between stats and Stephen King’s “Cujo,” the Secret Sauce ranks teams by how well they do in the three key areas that correlate to winning postseason games. As explained here by sauce-master Nate Silver, they are:
1. A power pitching staff, as measured by strikeout rate.
2. A good closer.
3. A good defense.
No. 5 DETROIT TIGERS
The Tigers have allowed the third-fewest runs in the AL, just 4.5 per game. Their strikeout rate is roughly average, with only starter Justin Verlander, who leads the league with 230 strikeouts, really jumping out in that department. The Tigers have been an average to slightly above average fielding club, with few standout performances (Clete Thomas has been strong in right field, though he can’t hit like a right fielder) but no truly poor ones either, and overall they rank in the top half of the AL in turning balls in play into outs. Desperation closer Fernando Rodney has blown only one save all year, but walks too many batters for comfort against strong postseason lineups.
No. 4 BOSTON RED SOX
We begin with Jonathan Papelbon. We continue to the staff overall, which is tied with the Yankees for the league lead in strikeout rate, propelled by Josh Beckett, Jon Lester, Daniel Bard and Papelbon. We conclude with a defense that has been surprisingly weak. The outfield has been defensively shaky, the infield has lacked a shortstop of any defensive quality until Alex Gonzalez came over, and Kevin Youkilis has had to play too much third base, not to mention a couple of scary games in left field.
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
1. Walt Davis Posted: September 13, 2009 at 12:01 AM (#3320579)How well do team pitching HR rates hold up in the playoffs? It's fairly amazing that the Rangers can be 2nd in RA in that ballpark without striking anybody out. Either their defense is spectacular (and under-rated in these rankings) or they don't give up HR. OK, they give up 1.1 HR/9 ... just like virtually every other team in the AL ... cursed rounding!
And the Yankees have the best defense in the AL? Really? I want some eyeball reports!
According to the article, the Yankees "rank third in the league in turning balls in play into outs." Which is to say, Defensive Efficiency. The team is now fourth by that measure, behind the Mariners and Rangers and essentially tied with the Rays.
Wait, that means it's wrong . . .
JUNK.
No need to look behind the curtain. All you need to know is that the Yankees have the best secret sauce in the Majors.
I'll give you Damon, although he's probably only below average, not bad. But Jeter has been average to above by UZR for 2 seasons now. There's no way he should be considered a bad fielder right now.
Also Swisher is, and has historically been above average in the corners, and Gardner is fantastic.
From what I've seen of Gardner, he covers a lot of ground, but Damon probably laughs at him as he tries to unload a throw to home plate on fewer than 5 hops.
;)
From what I've seen of Gardner, he covers a lot of ground, but Damon probably laughs at him as he tries to unload a throw to home plate on fewer than 5 hops.
;)
I've watched about half the games. Swisher is awkward, but he covers a good amount of ground. His UZR in RF/LF is in the +6-7 range in over 400 G's. He's +3 this year. He's a good fielder, despite his occasional misadventures.
Gardner can fly, he gets to everything. His arm is not strong, but nowhere near Damonesque, and is very accurate. In limited sample size, he has a strongly positive arm rating from UZR.
I think you're buying into stereotypes.
Nope. When I looked at what correlates with winning post-season series, to my surprise, offense didn't correlate at all.
This doesn't mean that he's a good defensive player. It simply means that he's had an above average UZR the last two seasons. Unless you can give a reason why he's better, you can't ignore the rest of his career, and that means he's an unequivocally bad defensive player.
But he has been a good defensive player this year. And unless they throw the clock back to 2007 when the playoffs begin, the Yankees will have a pretty good shortstop out there.
Agreed.
I do not share your possibly evangelical faith in current methods of analyzing balls in play, alas.
The work continues to improve, at least, imo.
"I think you're buying into stereotypes."
I think you're buying into the idea that UZR can't miss on any player, even to a degree.
Edit: Actually, maybe you were talking more about Swisher than Jeter.
I *have* watched Swisher play, and the fact is that while he's made some crazy routes to balls, I haven't seen him miss too many balls that a normal right fielder would catch. He'll probably be pretty awful in a few years, like Bernie became, but he seems okay out there right now.
Swisher is an odd duck defensively - not sure I'd completely trust any eye observation or defensive rating. I'll grant that sometimes the ball eventually ends up in his glove, but overall I'd have a hard time saying this is a "plus" defensive team overall - granting that they have won titles in the past 15 years with less....
in terms of defense, Damon (-8 UZR/150) combines inconsistent health (there are some games where he is noticablly slower than others) with inconsistent play (other than Nick Swisher, he hsa the most gaffe plays on the team.) Cano is at -5 but is currently going through a rough stretch. (he's amazingly inconsistent. going from one extreme to another quiet frequently) A-rod's at -6 but he was around -30 after his first month. so he's actually been quiet good since then. everyone else is either above average or right around average. with Gardner being awesome in his limited sample size.
Positioning effect defense... A LOT. so it is certainly concievable that a guy with the wrong positioning for a long while could end up looking a lot worse than his real ability.
Looking at individual UZR numbers for the last couple of seasons (without weighing and regressing, etc.), you'd have a hard time saying the Yankees are good defensively. And it's not all on Jeter, either. He's having a good year according to UZR, and it may not be completely a fluke/luck, as he was around average last season. But the earlier seasons aren't going away -- he's probably been at least a bit lucky this year. I suppose some people will want to blame Jeter for A-Rod's mysterious transformation from excellent SS to below-average 3B. I'd say it more likely age and injuries, but what do I know?
I can buy Teixeira being average or a bit above at best, but his glove at first is overrated, and UZR agrees. Cano was great in 2006, but pretty bad every other season (including this one). Damon's range is, well... approaching the quality of his arm. And why isn't anyone mentioning that Melky is average at best in CF? Posada is a great hitter. Not so much on defense, although there's no UZR for that at the moment. Whatever people think of his routes, in RF Swisher is arguably the best defensive player relative to position among the current starters, although if Gardner gets more playing time, that could really help.
For Swisher, he's been statistically a plus RF/LF for his whole career, this is not a one season fluke. He also looks to me like he has good range. Finally, several teams thought he could hack CF (he really can't) which says they must think his defense in the corners is pretty good.
Everything, except his occasional awkwardness/bad route lines up on Swisher being a good OF. I think it's far, far more likely that you are "missing" on him than UZR.
For Jeter, every observer says he looks much, much better at SS the last two years. Whether it is improved positioning, or health, he's gotten better.
Not necessarily. However bad he *is*, his ability to get to most balls is not zero, it's just lower than the average shortstop. A full season is a small sample size for defensive numbers. Just as bad hitters often have a good two months, a bad defensive player can have a good season.
I don't agree with Larry here. We're at almost two full seasons now, and the Stats and the eyes say he's much better. Why is it impossible to improve on defense, even at his age? Jeter was always a good athlete, so his poor defense was somewhat surprising. If it was a matter of technique or positioning, rather than physical, it is very possible that he finally figured it out, with some help from the new coaching staff. (I don't think it is a mere coincidence that his defense improved as soon as Torre left.)
Anyone in the slightest bit familiar with Prospectus Entertainment Ventures, LLC should not be in the remotest surprised by this, any more than you should be surprised the sun came up this morning.
Over his career, Jeter's range factor per 9 innings has been 4.16. He's been worse each of the past 3 seasons: 2007: 4.01, 2008: 4.05, 2009: 3.96. It is very hard to reconcile this with the idea that he has become a better fielder.
There are of course reasons that RF alone may not tell us how good a fielder a player is. Maybe Jeter has had fewer opportunities over his career than other SSs for various reasons. But how likely is that Jeter became a better SS in recent seasons, but coincidentally his opportunities (already very limited) dropped just enough to exactly offset that improvement? Far more likely that he's the same player, but 3 consecutive GGs have led many observers to "see" improvement that isn't really there.
Thing is, I don't disagree with what Prospectus Ventures does. It's smart. There's a reason BPro has been able to do what it's done and why it has the reach it has. It's not anti-BPro to point out a part of the business model. It's not snark when I point out, as I have done many times, that Prospectus Ventures decided years ago that the cash cow was going to be fantasy players. Regardless of whether or not their focus is different than mine, it's not my place to put a value judgment of right/wrong on that decision, outside of debating its long term potential (which I do have some issues with.) So I apologize for the miscast aspersion here, truly, but at the same time I do not apologize for my intentions when I posted it.
A lot of that stems from the time BPro had its lawyers send a cease-and-desist letter to BPrime, claiming that it was ripping off BPro's name. That did not go over very well here, and some of us have a long memory for that kind of thing.
There are of course reasons that RF alone may not tell us how good a fielder a player is. Maybe Jeter has had fewer opportunities over his career than other SSs for various reasons. But how likely is that Jeter became a better SS in recent seasons, but coincidentally his opportunities (already very limited) dropped just enough to exactly offset that improvement? Far more likely that he's the same player, but 3 consecutive GGs have led many observers to "see" improvement that isn't really there.
Range Factor? Really?
Back in the GG seasons, Yankee fans here were routinely panning them. The turnaround didn't come until 2008.
BTW, it is quite possible than the composition of the Yankee pitching staff changed in 2008-09 leading to fewer GBs to SS, at the same time Jeter improved the % of those GBs he was fielding.
A major contributor, off the top of my head, is Wang's effective disappearance from the rotation the last 1.5 yrs. Quick fangraphs check, the Yankees GB % is down to ~43% in 2009 from ~46% in the recent past.
2006 43.3
2007 42.8
2008 45.7
2009 42.5
Over Jeter's career, many pitchers have come and gone. The excuses also change. But one thing is constant: Jeter makes far fewer outs than other shortstops.
Yes, UZR says he's been passable some years. But plus/minus, based on same data, usually says he's terrible. All of these metrics have limitations. But given enough seasons, the difficulty of chances a SS has will tend to even out. And Tango controlled for the pitchers Jeter played behind -- and much else -- and still the verdict is always the same.
2006 43.3
2007 42.8
2008 45.7
2009 42.5
Odd, I got different numbers when I checked before, but now I see the same ones as you. Must be user error.
Correct. But "Jeter is awful defensively" doesn't (necessarily) follow from "Jeter makes fewer outs".
My argument is that Jeter's failure to make outs resulted from a combination of his defensive skills, the way that the Yankee pitchers pitched, and the way that the Yankees positioned their infield (at least under Torre), and that until you can figure out how to quantify the effects of the latter two factors, the best that you can do is conclude that there is no evidence of GG quality for Jeter's defense (which is probably damning enough).
Again, not necessarily. Infielders tend to do worse in zone-based ratings when playing behind a flyball pitcher than they do behind a groundball pitcher, to some extent because GBs hit against a flyball pitcher tend to be more difficult to field. If you have a SS who plays for years behind a staff of mostly flyball pitchers (as Jeter has for much of his career), I would suggest that you would probably NOT see the degree of difficulty evening out.
-- MWE
I'm skeptical of the positioning argument: if Jeter was required to play in non-standard positions given the hitter/pitcher/baserunner/out situations, it seems likely this was a response to Jeter's weakness rather than its cause. Different positioning would just as likely have led to still worse numbers for Jeter.
So no, I don't think we need any more data to conclude Jeter has been an historically bad SS (among those allowed to play a lot of innings there). The burden of proof now rests on anyone wanting to argue the contrary position.
I agree that Jeter shouldn't be considered a bad fielder right now. Why do we have to give a reason why he's better, apart from citing something like UZR? Maybe he's learned a new trick or two. While fielders generally decline in their 30s, especially SSs in their mid-30s, that doesn't mean that all of them have to. Jeter was so bad that if he could just learn something new (say, with respect to positioning), there were a lot of gains available to him.
Barry Bonds changed his game at age 35. I don't see why Jeter couldn't have.
And citing Range Factor is weak sauce, GuyM.
2009 is the first time in a while that the Yankees have had multiple left-handed starters.
Maybe there's a lot of transparency on the metrics (where "a lot" means "not as much as everyone else but not entirely nonexistant.") Maybe there is "some transparency" to FRAA (I can't find anything useful to figure out how it's actually computed, though, and I've checked all the expected places - old annuals, Google's USENET archive, etc.) I'll leave that point alone.
But there is very little accountability.
What we do know about FRAA is that it simply isn't a very good fielding metric, especially restricted to a single season sample size. We do know that there are better fielding metrics. Dan Fox even came up with one when he worked for BPro. And yet FRAA persists.
As for WXLR, Tom Tango has addressed the problems with the underlying Leverage metric BPro uses.
So then we get to the Secret Sauce. Again, it's not that nobody knows what's in the sauce; we all know it's Thousand Island dressing. It's that there's no evidence it works.
This doesn't mean that [Jeter]'s a good defensive player. It simply means that he's had an above average UZR the last two seasons. Unless you can give a reason why he's better, you can't ignore the rest of his career
It doesn't matter if he "is" a good defensive player. He's been playing well (or at least, decently) defensively. No one has to give a reason why. What he did in 2000 has nothing to do with now.
I'd certainly pick Posada as a bad defender over Jeter these days.
He's been worse each of the past 3 seasons: 2007: 4.01, 2008: 4.05, 2009: 3.96. It is very hard to reconcile this with the idea that he has become a better fielder.
It's very hard to reconcile how he looked in 2007 (incredibly horrible) with how he looks this season (quite competent).
2009 is the first time in a while that the Yankees have had multiple left-handed starters.
Which should mean more balls to SS, not less.
[Gardner']s arm is not strong, but nowhere near Damonesque, and is very accurate.
I actually think his arm is average strength, but not accurate at all (i.e., too many of his throws tend to go straight down, rather than forward).
Cano was great in 2006, but pretty bad every other season (including this one).
I think he's been about as good (i.e. pretty good) this season as he's ever been (certainly better than '05 or '08, when he had lots of weird "mental" errors). My friend thinks he's been excellent this season.
So, all of this goes to show..........we still don't know #### about defense!
I'm a moron.
First of all, from the definitions of "defensive games" and "expected outs," exO should be a constant multiple of DG, namely the average number of putouts per game at that position. Sure enough, looking at Jeter's page, the ratio is about 2.6 for every year from 2002-08. But for 2009 the ratio is only 1.9. What gives?
It's hard to figure out how many plays Jeter actually made looking at only putouts and assists. (Maybe PO + A - DP is a good approximation?) But for an outfielder, plays made should be almost equal to putouts. Take Grady Sizemore as an example. In 2005 he had 373 putouts to 348 expected outs. His UZR was +5.4. In 2006 he had 409 putouts to 428 expected outs. His UZR was +14.8.
I can't figure out what is going on. It shouldn't have to do with the difficulty of his chances, because that should already be accounted for in the calculation of expected outs.
By the way, Sizemore also has the weird issue where the ratio of expected outs to defensive games is much lower in 2009 than in previous years.
Different run value of the catches and misses? Deep catches would save more runs per play. Not sure if that's actually what the issue is or not.
Does seem unlikely. The difference between a single and a double is what, .3 runs? Yeah, turning -19 plays into +15 runs would take a whole lot.
Steven, if I'm not mistaken, and please correct me if I'm wrong, the particulars of FRAA / PECOTA / (the adjustments for) WARP2 and 3 / BPro's WARP / PAP are all kept back. If that's the case, we'll have to agree to disagree on our assessments of the overall transparency.
No one is saying "good" we're just saying not "bad". All the evidence (UZR, observation, and now Fielding Bible) seems to indicate that while Jeter was "bad" (~-15) three or four years ago, he has stepped up his game defensively and is now average (~+/-5).
FRAA, the timeline adjustments - those are still pretty secretive.
You must be Registered and Logged In to post comments.
<< Back to main