Yes, I meant only at the player level. But I think we agree that a SS who is, say, truly -20 runs -- and plays alongside average 2Bmen and average 3Bmen -- will only be about -12 in TZ. We don't disagree about that, do we?

Looking at park adjusted DER is an interesting idea. Probably would need to do it separately for IF and OF. It would help establish a reasonable overall distribution, but probably couldn't be precise enough to establish the really extreme players like Jeter, Ozzie, Belanger.

I do think that WOWY has a potential problem in the OF in terms of "ball hogs." For example, I have a hard time believing that Andruw Jones has really been 200 plays better than Mike Cameron. I'd have to be sure that the Braves corner OFs (and middle infielders) didn't have unusually low PO totals before I'd be willing to accept that. In the IF, though, this isn't usually a huge problem.

Gonfalon Bubble FTW.

Does that PO/A adjustment take into account the possibility of a long-term situation where the defense has significantly more/less BIP than league average over 10-15 seasons? Because that's what you had in STL during Ozzie's reign: about 1500 more balls in play than league average.

Conversely, while Jeter has been with the Yankees, they've had nearly 1200 fewer balls in play than league average.

That's about a 2700 BIP swing between the two teams per their league averages over similar time spans, or about 175 BIP per season.

Won't make Jeter come close to being Ozzie under any circumstances, but it does seem relevant in terms of the methods being bandied about.

Back onto the thread topic: I'd love to see JP's lists broken down into two: one for 2000-04, and the second for 2005-09. I think that would be a more useful ranking method than the "decade" rubric. While Rob finally got out of his plaid shirt, he's been slower to jettison some of his other forms of mental typecasting.

If you're going there, why is 2000-04 any more useful than 2003-2007?

If you're going there, why is 2000-04 any more useful than 2003-2007?

Because very few players have 10+ year runs of excellence?

I think you should do Don's suggestion one better. Give me a list of the best players from 2000-2004, then a list for 2001-2005, and so on through the decade. Then graph that so that I have a nifty chart of how players rose and fell, and what players took their places along the timeline.

If you're going there, why is 2000-04 any more useful than 2003-2007?

last 5, versus previous 5?

If you're going there, why is 2000-04 any more useful than 2003-2007?

Yeah, I don't think Don would argue that it is.

The problem with the ten years, of course -- well, aside from the fact that this particular ten-year period is arbitrary -- is that someone who played all ten years has a huge leg up on someone who didn't. That basically queers the rankings and makes this more of a trivia thing -- so heavily dependent on playing time -- than it is any useful or even very interesting analysis.

Trimming the time period down to five years would perhaps be more interesting, though still fraught with the same endemic problems.

I'd be more interested in seeing a "best ten years of the past quarter century" or whatever than this capricious whimsy of a time period.

I think that would be a more useful ranking method than the "decade" rubric. While Rob finally got out of his plaid shirt, he's been slower to jettison some of his other forms of mental typecasting.

Don, I wonder if an editor mandated that Rob create this list, rather than him deciding to do it on his own.

Don: Sure, it makes sense to adjust for BIP. If you know GB/FB ratio, you can adjust for that. But let's put it in perspective: 175 BIP probably means a difference of about 20 expected outs for the two SSs. So let's narrow the Jeter/Ozzie gap another 280 plays, or 210 runs. So we've got Ozzie's advantage down to "only" 900 runs, or 90 wins. And we're running out of things to adjust for...

Ray,

You did a good job of representing most of the responses I would have made had I made them. :-)

I think Sam's idea is pretty good. Your concept is interesting, too.

If Rob wasn't such an inveterate listmeister, I might think that he was ordered to do this. But he can come here and tell us if that's the case, and then I will rescind the snark. He's also thinking about the MSM audience when constructing that subelement of his material, though the comments from his readership weren't too kind on this one, and they might've been <u>twice</u> as unkind had he created two such lists.

A brief look at how the half-decade rankings change is not anywhere close to definitive in terms of ranking or establishing "value" (however far you're taking that task), but it's interesting if only to see how dynamic the lists actually are. It's probably similar across any two half-decade rankings (for example 95-99 and 00-04, 00-04 and 05-09). Looking at the latter comp (only at the hitters, and using OPS+ because it's handy), however, we see that there is a lot of movement:

<u>Player, 0-4, 5-9</u>
Bonds, 1, -
Giambi, 2, -
Pujols, 3, 1
Manny, 4, 4
Helton, 5, 33
Thome, 6, 25
Edmonds, 7, -
Sheffield, 8, 58
Sosa, 9, -
Guerrero, 10, 16
Delgado, 11, 27
A-Rod, 12, 2
Giles, 13, 51
Berkman, 14, 5
Thomas, 15, -
Abreu, 16, 48
C. Jones, 17, 3
Walker, 18, -
Rolen, 28, 85
Cabrera, -, 6
Ortiz, 44, 7
Braun, -, 8
Howard, -, 9
Teixeira, -, 10
Hafner, -, 11
Fielder, -, 12

(-) means fewer than 1800 PA in the five-year period and not ranked

This just shows volatility over a decade. What's interesting here is that the number of hitters who had a 140 OPS+ decreased from 18 in 00-04 to just 12 in 05-09. Why did that happen? Whatever the reason, it's why Manny ranks #4 in each half despite seeing his OPS+ drop from 167 to 153 over the respective periods.

It would be more interesting--and more problematic--with pitchers added in and whatever fielding component that people want to superimpose. Still would be interested in JP's list done this way--or Sam's way, or yours. Deciding before one looks at the list that it's going to produce trivia seems premature, even if the likelihood of that being the case seems high...

fantastic list, even with debateable discrepencies...not sure if I rank Pujols ahead of Arod personally, unless I allowed personal feelings to figure into it... I think Pujols is the ultimate professional in almost every aspect of the game, including preparation, but I know that as a local fan, I don't get exposed to other players at the same level, arod could be arguably the same quality guy(even with the negative press)

Yes, I meant only at the player level. But I think we agree that a SS who is, say, truly -20 runs -- and plays alongside average 2Bmen and average 3Bmen -- will only be about -12 in TZ. We don't disagree about that, do we?

I agree on that, as long as they are hits into the outfield. Infield hits and errors are charged 100% to the SS. A SS who misses an average mix of 20 plays would be about -14.

This story just keeps on giving. ESPN bumped it up to the front page last night, the 2nd or 3rd promo box under the lead picture. The comments have resumed. My current favorite.

i dont care what fancy calculations you are using - any list that ranks jd drew ahead of mariano rivera deserves zero credability and was done be someone who never watched a live baseball game. i'll tell you what, why dont you look at it this way stats boy. if in 2000 you held an open draft for ALL players AND you had the knowledge you have now, who would go where

I'm sorry, I'm sorry... but one more to share

yes...this guy neyer can't even spell baseket ball...a frat boy getting an East Coast job like little Georgie ol' Bushie jr....how pathetic

????? - but it made me laugh

Mark Bellhorn is a notable omission.

This is just from 2002 on and obviously using BIS' UZR info from fangraphs...

Jeter exO/INN at SS is .270 in 10238 innings.
Other Yankees SS have been at .273 in 1331 innings.

AL teams during this time other than the Yankees have been at .289, for a difference of about 27 expected outs per season (1450 innings).

If there's a scoring bias in favor of Jeter on BIS, wouldn't you expect other Yankees SS to be much closer to the league average? Doesn't it seem likely that, as I believe Mike E. found years ago, the Yankees do have an abnormal BIP distribution?

Assume the difference between Jeter and other Yankees SS is scoring bias, that means Jeter should have had about 28 more plays counted against him, which would be (trying to remember Dial's run values) about 21 runs. That'd move Jeter to about -7.5 runs per 150 games during this time.

To get to the -30 that WOWY has for Jeter, you have to believe that the Yankees do have a league average distribution on BIP, and I guess the scorers for BIS are just not counting them against him or the other Yankees SS for some reason. I guess that's possible, but I have a hard time seeing them 30 plays off every year on one team's position.

A little project recently at Baseball Fever came up with this list:

1 1B Albert Pujols
2 3B Alex Rodriguez
3 LF Barry Bonds
4 SP Randy Johnson
5 LF Manny Ramirez
6 SP Johan Santana
7 3B Chipper Jones
8 SS Derek Jeter
9 RP Mariano Rivera
10 1B Todd Helton
11 SP Roy Halladay
12 RF Ichiro Suzuki
13 CF Carlos Beltran
14 SP Pedro Martinez
15 RF Vladimir Guerrero
16 LF Lance Berkman
17 3B Scott Rolen
18 RF Bobby Abreu
19 CF Jim Edmonds
20 C Jorge Posada
21 2B Jeff Kent
22 1B Jim Thome
23 1B Jason Giambi
24 SP Curt Schilling
25 SP Roy Oswalt

Kyle:
I never suggested that the problem with the BIS data, if there is one, is Jeter-related. All I know is that it's inconsistent with a lot of other evidence.

I agree that Jeter had fewer than average opportunities, if only because the Yankees have had a lot of high-K pitchers. But WOWY effectively adjusts for that, and still finds Jeter to be hundreds of runs below average.

In any case, Jeter has 421 assists per 1450 innings. The average AL SS is around 465. Even if you believe that fewer opportunities explain maybe 22 of those assists, he's still -23 plays a year, or about -17 runs. That's still atrocious.

Even if you believe that fewer opportunities explain maybe 22 of those assists, he's still -23 plays a year, or about -17 runs.

How does -23 plays at SS, where a missed play is almost always a single, add up to 17 runs? For whatever reason, I thought the estimated run value for a single was a lot lower than 0.74 runs.

For whatever reason, I thought the estimated run value for a single was a lot lower than 0.74 runs.

It is, .46 runs in Palmer's LWTS, but that's compared to average. If the fielder had gotten to the ball it would have been an out, and an out is around -.25 runs.

I never suggested that the problem with the BIS data, if there is one, is Jeter-related. All I know is that it's inconsistent with a lot of other evidence.

I know it sounded like I was implying that, but I did not mean it that way. I should have added scoring error as a possibility rather than just bias, and I'm ignoring other potential errors on their end since we're talking just about Jeter right now.

I agree that Jeter had fewer than average opportunities, if only because the Yankees have had a lot of high-K pitchers. But WOWY effectively adjusts for that, and still finds Jeter to be hundreds of runs below average.

Does it even control for GB/FB rates, though? I still don't see how assuming what the BIP distribution is could ever be better than using actual BIP data.

In any case, Jeter has 421 assists per 1450 innings. The average AL SS is around 465. Even if you believe that fewer opportunities explain maybe 22 of those assists, he's still -23 plays a year, or about -17 runs. That's still atrocious.

Atrocious and still not close to -30 runs that was talked about earlier. I just don't think there's any way you're going to convince me that using WOWY even comes close to being as accurate as BIP data.

I asked SG in ATL for Jeter's ZR runs since 2002, and it ended up being -50 for about -7 a year. I would prefer UZR since I'm sure ZR has the same problems total zone does in underrating bad defensive players, but it does at least use Stats data and I see no way you can get close to the WOWY numbers still.

You'll never see me argue that Jeter is anything but a bad defensive SS. It's just that the WOWY numbers don't seem believable at all, and when I see fielding stats with superior methods (IMO) back that up, I'm not going to change my mind.