I believe the Joe Torre era is in its final hours.

Amazingly, with defense and attitude included, Hillenbrand (this year) may actually be a worse option than Cairo!

Ah, remember the halcyon days of c. 2000, when talking heads actually compared Hillenbrand and Soriano?

1 A large part of me hopes so. A small part of me wonders if Cashman isn't going to hold out, try to get a big bat - I'm not sure who - and see if things will right themselves, a la 2000. It wouldn't surprise me. But I don't see both a David Justice-type bat and (more importantly?) a Glenallen Hill-type bat available.

2 Rivalry-fueled hype, maybe? They weren't comparable then, and aren't comparable now. Signing Hillenbrand and expecting anything positive could be on the level of the T-Ball Long signing of last year. I suppose Cashman has to try, when the only other option Torre will use is Cairo.

2 I do indeed remember those days. Man, that was a long time ago.

I seem to recall that the comparison centered around the fact that neither one ever met a pitch he didn't like.

3 Oh my God. We did have Terrence Long, didn't we? The horror, the horror.

And Kenny Lofton, too. Remember Kenny?

And we already have a David Justice-type bat. His name is Bobby Abreu and he came over last year. Of course, DJ was great his first year and lousy his second too, although DJ had injuries as an excuse, iirc.

Good times...

"Tonight the Yankees look to continue their philanthropic tour of the gloomy gusses of baseball. "

Pure poetic genius, Cliff. Hats off.

We play today, we get humiliated today.

Q: Do you think a team with Mariano Rivera performs better or worse in 1-run games?

A: Since 2001 the Yankees have been worse in 1-run games than they have been overall.

NYY win % in 1-run games since '01: .561
NYY win % in games decided by > 1 run: .602
NYY win % overall win % since '01: .591

The Yanks have played 271 1-run games since '01 so it's a fairly large sample. I just think it's interesting that the team with the best reliever in history and one of the top 5 relievers each of the past 6 years is WORSE not better in one-run games.

Any ideas on how the NYY could be worse in 1-run games with Mo on the team?

7 Good teams (Win% .591 = 95 wins/season) tend to score lots of runs and give up few runs, resulting in many lopsided victories and few lopsided defeats. The games that they tend to lose will be close--indeed, they will be in most of their games even if they lose.

Anyway, the real question is not to compare the Yankees' Win% in 1-run games to themselves, but to the rest of the league. Is the teams 1-run win % (.561) significantly higher than other teams'--that might say something about the relationship between the greatest closer of all time and winning close games.

7 They lose a bunch of one-run games with Mo watching from the bullpen, waiting for the save which never comes?

8 Here are a few AL teams for comparison.

NYY win % in 1-run games since '01: .561
NYY win % in games decided by > 1 run: .602
NYY win % overall win % since '01: .591

LAA win % in 1-run games since '01: .527
LAA win % in games decided by > 1 run: .557
LAA win % overall win % since '01: .549

BOS win % in 1-run games since '01: .540
BOS win % in games decided by > 1 run: .581
BOS win % overall win % since '01: .570

MIN win % in 1-run games since '01: .570
MIN win % in games decided by > 1 run: .548
MIN win % overall win % since '01: .553

Any thoughts? I still think that the Yankees should 'close out' a far greater percentage of 1-run games where they are leading because of Rivera. San Diego has a big advantage in 1-run games with Trevor Hoffman. Why doesn't NYY with Mo?

SDP win % in 1-run games since '01: .546
SDP win % in games decided by > 1 run: .462
SDP win % overall win % since '01: .486

What are some possible correlations I could run with the data? Save rate of closers? Team defense? I wonder why there is such a variation in how teams do in 1-run games.

4 Ah, but Kenny Lofton had something to contribute. Instead, Teaman kept starting Bernie.

1 As we realized yesterday - If Joe sticks, Cashman has cover. I suspect Joe sticks through October. I hope I'm wrong.

11 I think you overestimate Lofton's potential. At that point he was a better defensive outfielder than Bernie, but not a good one. And he wasn't a better hitter.

Prediction: We trade Shelly Duncan to LAA for Hillebrand. Duncan catches lightning in a bottle and hits 15 HR in the second half of the season. Hillebrand, along with Kyle Farnsworth leads the league in breaking bats in the dugout.

10 i think 8 gives you the answer stated very simply and correctly. 1 run games can be 1 run victories or 1 run losses. it isn't all on the closer. if the yankees are down 8-7 going into the 9th, mariano probably never even pitches. in fact, even if it's 8-8 going into the 9th he sometimes doesn't pitch, and they lose by a run.

12 That's the point. Lofton should have been the CF. Bernie would have been the DH.

Indeed, I'd argue that Lofton has been more valuable over the last four years than Bernie. And Lofton is still contributing an average CF - bat and glove - at age 40.

One-run records tend to snap back to .500, not to a team's overall winning percentage, thus playing .561 ball in one-run games is an impressive mark likely due in large part to Rivera, not a low mark achieved despite his presence.

Also, the issue in 2004 wasn't so much Lofton (.275/.346/.395) vs. Bernie (.262/.360/.435), but Lofton vs. Ruben Sierra (.244/.296/.456).

Somebody wake up Cashman and tell him to fix the team.

DL Damon - call up Duncan.
Play Duncan every day at 1B for two weeks.
Send down Nieves - call up Omir Santos at C.
Send down Bruney, call up Britton.
Trade Myers for anything, call up Henn
Trade Villone for anything, call up Edwar Ramirez
Mix and match ARod, Jeter, Abreu, Posada and Matsui at DH
Play Santos 1 day a week
Play Cairo 2 days a week
Play KT 3 days a week
Always bring in Mariano in tie games in the 9th, 8th if rested
New rule - no RP can pitch on consecutive days except Mo
New rule - Miguel Cairo can never bat higher than 7th
New rule - Jeter bats leadoff

I can live with the Yankees losing but I'm damn tired of seeing the same stupid flaws cost us again and again. Go young, go hard, go smart. I can live with that kind of team but not what I've been seeing. It's been a sad run.

10 Cliff beat me to it 16. But let's look more closely at your numbers--SD [w% .546] has not had a bigger advantage in 1-run games than the Yankees [w% .561]. Moreover, if you look at the cross section of teams that you present, there is little pattern overall other than the Yankees seem to have demonstrated excellence in both close and not-so-close games.

I am willing to complain about Torre's use of the BP, and I may agree that it has cost games. But, the numbers you present don't add much to the case, nor prove that his mismanagement is worse than that found on other teams.

Cliff - I'm with you on 2004.

DH Bernie > DH Ruben
CF Lofton > CF Bernie

And of course, Lofton would have been the CF they needed in 2005 (instead of the Bernie/Womack/Crosby monster).

See, the problems with Yankee management have existed for a long time. It's the reason they lost in the postseason for the last 5 years (2002-2006): Poor replacements for players lost and mismanagement of the players they had.

17 I'm not sure why Bruney deserves to get sent out before Farns or Vizcaino. He puts runners on, like them, but he's much better at keeping them from scoring.

19 Did Tony play CF? I thought he was more at the corners but my memory may be hazy.

Man, you know its going bad when you think back to the days of Womack as the good times.

Can any reporter please take Joe to task for bringing in Mo last night to "get work in"? as opposed to getting that work in, in a tie game.

Or hell, even just forgetting which days Mo has pitched on.

Kenny Lofton in CF:
2002: 105 RATE
2003: 104 RATE
2004: 105 RATE (Yankees)
2005: 110 RATE
2006: 93 RATE
2007: 90 RATE

Bernie in CF:
2002: 89 RATE
2003: 93 RATE
2004: 90 RATE
2005: 97 RATE
2006: 78 RATE

22 Why would they do that? That would mean doing their job but risk losing their access.

23 Wait, has this team gotten so bad that we are just back to debating Lofton vs. Bernie?

What other fun debates can we cook up? How about Mantle vs. DiMaggio for CF?

25 Yup. More interesting to me than the current variety.

Besides, the point was: "[T]he problems with Yankee management have existed for a long time...Poor replacements for players lost and mismanagement of the players they had."

The difference is: The more players that have had to be replaced, the worse the team has gotten. The GM has continued to plug huge holes with uncreative (Damon) or downright dumb solutions (Womack, Long, Cairo, Nieves). And the more power the GM got, the worse still the solutions. Everyone rails against Tampa, but if more people were involved in the decision making, Cairo would not be the starting 1B.

25DiMaggio in CF:
1947 91 RATE
1948 98 RATE
1949 107 RATE
1950 92 RATE
1951 102 RATE

Mantle in CF:
1952 97 RATE
1953 97 RATE
1954 98 RATE
1955 102 RATE

Clearly Stengel made the correct call in 1952. Also, any statistic that rates either Mantle or DiMaggio as average is worthless.

25 Meanwhile, go look at what other options have been out there.

Look at what Toronto gave up for Overbay - it was the Randy Johnson pu pu platter.

Look at what it would have cost to sign Huff and Gomez and Lieberthal for a respectable bench.

Look at the relievers in the Yankee system.

That's 1B, the bench, and the bullpen. The future wasn't forsaken. The money spent is less than Pavano's contract.

27 Why, cause you saw Mantle and DiMaggio play with your own eyes?

Suffice it to say - they weren't outstanding because of their gloves.

29 What difference would it make if 27 saw them play in person? A valuable metric should not need personal forensic observation for it to work, and indeed it should be able to remove personal and anecdotal bias from the estimation of defensive prowess ("I saw the Mick make a great catch--I know he was really good in CF").

Yet EVERYTHING I have ever read by anyone who saw DiMaggio and Mantle play agree that they were excellent CFs--especially DiMaggio--at least before injuries took their toll on both. So, what to believe when evluating the defensive abilities of players from bygone eras?

Oh dear lord, I've created a monster :)

Someone needs to write the biography of Brian Cashman.

How an intern rose to be GM of the most powerful organization in sports.

How he lasted for more than ten years.

How he managed to consistently deflect blame without appearing as doing so.

How he wrestled control from the Tampa cabal and left the impression that all problems were of their making.

How he fails to see the importance of the bench.

How he thinks singular positions can be sub-par because the rest of the team is spectacular.

How he tried to manage away from Torre's weaknesses and toward his strengths.

26 More people involved in the decision making? The real problem with Cash has been his ridiculous deference to Torre...hence we have the tragedy of Cairo, the second coming of Donnie Baseball in Joe's eyes. For fear of having to work with Piniella (who apparently feels the same way about him as Zimmer publically stated...Cash is not a baseball guy and, ipso facto, brings nothing to the table), Cash urged George to let Torre finish his deal, thinking he could Torre-proof the roster. Well, the injuries to the rotation scotched that quickly and the bullpen over-use doubled down the damage. The rest is the usual lunacy from Torre and, as much as I like Cash, the buck really stops with him. One of the few sane things George has declaimed. Sorry, Cash, you screwed the pooch.

31 Don't apologize-- I was hoping for a topic like this to lighten the mood and change the discourse for a day or two!

How is RATE calculated, anyway? Or is it one of the secret stats whose formula is hidden?

30 And popular opinion of today would make Jeter an outstanding SS. Actually - that's the comparison. Jeter's bat at his position makes him far more valuable thn the players that came before him. Same deal with Mantle and Dimaggio - except their bats were that much better.

Me, I'll trust the stats over what newspapers and fans say every day of the week. Cause the former summarize everything while the latter have selective memories.

And the stats don't say Mantle and Dimaggio were butchers. They say they were average with some very fine seasons. I can certainly deal with that, especially since their bats were ridiculous. As CF's, they are extremely valuable in the history of the game. The only player since that comes close is Griffey Jr, and then his bat doesn't quite measure up.

35 But Jeter is a great fielding SS. Joe Morgan said so and he's the top analyst in the world. DUH!

The irony in all of this for Cash is...I betcha Torre doesn't respect him either. It would have been interesting to see how Torre's tenure would have gone if Gene Michael had somehow survived George's purge. Apparently, Michael is not enamored of Torre's managing style. Again, Cash's biggest mistake was to not fire Torre after 2004 and install his own man. He could have had the best of both worlds: independence from Tampa and someone in the dugout who shared his vision (not sure if he really has one but whatever).

Hey guys look at the bright side. We're not Real Madrid. They just won the league title for the first time in a few years, then promptly fired their manager.

33 It's like Wikipedia and the wisdom of crowds. You aren't going to get masterpieces, but you tend to avoid idiotic solutions.

Me, I look back at when Tampa was involved, and it's just as hard to parse the influence of George on them as Tampa on Cashman.

So while people wory about Tampa (myself included), I realize now that was a fantastic political machination by Cashman to cover his own ass.

35 But even the Jeter comparison fails. We have endless reels of tape that can be pointed to which highlight his defensive weakness (lateral range) as well as his strengths. The constant drumbeat by those citing stats has, finally, started to counter those who praise Jeter's defense. But with players from previous generations, we just don't have the kind of visual evidence to hold up against the defensive stats, many of which are still controversial (and not just by those eschew them, but in stat-friendly baseball circles).

35 "Me, I'll trust the stats over what newspapers and fans say every day of the week. Cause the former summarize everything while the latter have selective memories."

do stats really summarize everything? does a new, vague metric really capture the defensive prowess of fielders from 40-60 years ago? i don't think so. say what you will about offensive stats, most of them are relatively objective. defensive stats leave much more to judgment, don't they? how do we know the RATE of guys from way before the stat was invented? is the footage that good?

even offensive stats don't tell everything. i think we should push for a whole new system, a whole new way to spend the insane amount of money baseball makes. each stadium should have sensors all around that measure the position of the ball at all times, the position of the fielders, etc. and the ball itself should have a sensor that measures the velocity, spin, etc at all times. same with the bat. this way we could actually record "everything" and derive stats from that, instead of creating more and more stats to try to recreate "everything."

39 "It's like Wikipedia and the wisdom of crowds. You aren't going to get masterpieces, but you tend to avoid idiotic solutions."

Which would seem to contradict your thesis about defensive reputation, where popular wisdom and the opinion of the masses is often wildy misinformed!

42 you beat me to that

41 Bad wording on my part. But the stats have a better chance of being right than the media or fans. They will always take more information into account and do a better job of parsing it. The human mind just isn't built like a database.

The point is: Just because some geezer said DiMaggio was fantastic, I shouldn't believe it. Because a bunch of geezer already say the same thing about Jeter.

I'd buy your proposal. And with RFID chips, it's possible today. You would never need an umpire again. Good luck convincing MLB.

44 "You would never need an umpire again." that's one of the goals.

the other big thing that needs to change is that the cf camera should actually be from straightaway cf, not left-center.

Yes, RATE is one of those secret stats. I think it's one where they chart every batted ball - which makes it subjective to begin with. Then they make assumptions that we're not privy to. Maybe it's an excellent stat, maybe it's not. But just because it's a stat doesn't mean that it's better than observation, not unless we know how the information's derived.

You can find lots of testimony today that says Jeter isn't a great shortstop, and it's not just on the web. Find me one, just one citation that says that either Dimaggio or Mantle in his prime was an average center fielder.

And since RATE can't chart those balls, how is it useful?

42 Failures of crowd intelligence:

Too emotional
Emotional factors, such as a feeling of belonging, can lead to peer pressure, herd instinct, and in extreme cases collective hysteria.

http://en.wikipedia.org/wiki/The_Wisdom_of_Crowds

Jim, the Yankees won the division with Bernie in center and Sierra at DH. Do you think that having Lofton would have made the difference in the ALCS?

I'm not arguing about Torre today, because I agreee that he's no longer good for this team. But those who bash his past have to deal with 11 consecutive division titles somehow. They weren't just pushbutton titles, either.

45 If the cf camera were in dead center field, it would give you a great view of the pitcher's back.

49 I thought the angle they experimented with at ESPN a few years ago wa sgreat (dead CF but elevated); apparently I was in the minority and everyone complained, so the got rid of it.

48 we all know the success the team has had under joe torre. but we all also know that it's at least possible that other people who are no longer with the team had a big influence on torre's in-game decisions. and we also know that in 03 and 04 (not to mention 05 and 06 but they weren't as blatant) torre made some inexcusable mistakes that led to defeat.

50 exactly. it was so much better. they still use it when they show their k zone or whatever (maybe that's only fox now?). it's a much better angle, and it's a mistake not to use it.

In much happier (and much closer to real life) news, Humbug's Score Bard just became a dad again.

Congrats Ken!

44 Stats are only better if they're good stats. You seem to treat all stats as if they're purely descriptive stats, and they're not. BA, SA and OBP are descriptive and involve no interpretation. Most of the more complicated stats involve assumptions, interpretations and choices that make them something different.

In order to know if they're good or not, we need to know (a) how they're derived, and (b) if they accurately reflect what they claim to represent.

53 Please, don't lay the Yankees' success at the feet of Don Zimmer. And Torre obviously did a whole lot of things right in '03 and '04. I think it's entirely presumptuous to blame the losses on Torre while denying him credit for the wins.

But back to Hillenbrand and/or Bradley. Yes? No?

46 Find me one SABR member doing those studies in 1947-1956.

Any statistic is going to have shortfalls. In RATE, I picked one that seems to do a fine job.

And the SABR crew is always trying to do better. The stats will always be better than any collection of fan opinions.

55 The reverse is also true. Over the last six years, the mistakes and losses have overwhelmed the wins. And it's not 11 straight division titles. That's the kind of revisionism our president would not countenance.

48 If I may speak for Jim (with whom I often disagree!), the problem is not that the team won the division, or should have done better in the play offs, with Bernie in CF (for example). I think Jim is pointing out that the organization has turned a blind eye to fetering problems at key positions for years, basically getting away with it because they get super production from C and SS, or because the Sox collapse, or because some miracle shows up (like Aaron Small).

This has led to organizational complacency: 'we won the division last year with Bernie in CF, so why make changes.' Sports teams should always be looking to improve on the field, as long as the price is right. That doen't mean making wholesale changes every year, but there have been clear places for imporvement on this team for years, and they (seem to) have been ignored. Now, everything is happening all at once and they are in free fall.

But if 1B or bench or even BUC had been addressed more seriously in the last couple of years, there is a good chance that team would not be facing such a crisis now.

By ignoring problem spots for years, they created a team with a very slender margin for error, despite the massive star power in the starting line up. So, when someone says "it's not Cairo's fault they are losing, it's everyone else who is playing under potential, " they miss the point. By creating such a top-heavy team of aging stars, while ignoring festering weaknesses, the organization created a team uniquely unable to weather off years by the stars.

57 There, you're just wrong. Stats will not "always be better." They're only better if they're better. Just because they're stats doesn't mean they're better.

And no, there were no SABR members charting balls in the 50's. Which is why stats that involve charting balls are useless for that period.

55 have i ever said he doesn't deserve any credit for the wins? but you want to give him complete credit for winning all those division titles when, like i said, it's at least possible that he was heavily influenced by a better tactician. when he let gordon walk a guy and THEN give up a hit to make it 1st and 3rd with no one out after giving up a huge bomb to ortiz in game 5, before finally going to rivera who then pitched the 2 full innings, that was inexcusable. if it was possible that in the worst case scenario mo could pitch 2 innings, he had to come in after that home run. and after the walk, what was he thinking?

monkeypants - Shouldn't better teams have a better record in 1-run games? Rivera and Hoffman have been like clones over the past few years but somehow San Diego does 12% better than their overall record in 1-run games while the Yankees do 5% worse.

Jim Dean - Bruney can stay. I just thought a stay in AAA might shock him into throwing a few more strikes. I think Bruney is doing a decent job but he could be a lot better if he threw more strikes. I didn't mention Farnsworth or Vizcaino because I didn't want to get into trading big salaries or players with value. I don't think Myers or Villone have much value.

48 He would have helped more. Lofton was a better defensive option, better on the bases, and about equivalent with the bat. He offered more.

54 "In order to know if they're good or not, we need to know (a) how they're derived, and (b) if they accurately reflect what they claim to represent."

No doubt. But they'll still do better than the main stream media and popular opinion.

55 Torre has mismanaged many games - post-season included.

It's the same tendencies now. Just more decisions.

58 Of course, you're right. 10 titls in 11 years, 9 straight. That still makes it hard for me to say that the mistakes have overwhelmed the successes for the last 6 years. If you believe that the only purpose of a manager is to pull a few strings for two weeks in October, then you'd be correct. I think the previous 6 months count for something as well - and that a manager's most important contributions come off the field.

Of course I think there are a lot of things that should have been done differently. The failure to plan for Posada's aging galls me no end, and many of the other comments have validity. Many of them are also far less important than they're being made out to be. I just think there's a tone of condescension towards Torre that's entirely unwarranted.

OK, I give up. People here just know better. I done with it. Bye.

57 "The stats will always be better than any collection of fan opinions. "

I'm a stat guy, but this is just not true. If the statistic is seriously flawed, then we are better off with individuals' testimony (which can also be interpreted empirically).

The problem of stat-triumphalism is that it ails to recognize, as 54 points out, that may stats (at least the defensive variety) have subjective components built into them. They give the false sense of objectivity, when all they do is codify and quantify some indivual's or groups subjective opinion.

It's like the BCS. People think 'the computer' decides who's champion. That's bogus, it's people subjectively decising what weight to give to various factors, including how much to count polls (the single biggest component, and purely subjective).

62 You're confusing the issue by assuming a team should do better in one run games than their 'regular' record. The Yankees have been better in one-run games than the Padres over the period you analyze, as your stats show.

59 I couldn't have said it better myself.

I would also add that one reason the Yanks of 1996-01 did so well is that in 20 years we'll see they had three HOF players (Jeter, Mo, Jorge in 1998), one just miss HOF CF (Bernie), and very good starting pitching. Does any other team of the free agent era feature three HOF in their prime?

60 I'm fairly certain RATE does not involve any charts. Help please?

62 I agreed with Myers and Villone.

What contributions off the field are you talking about? His charitable contributions? His commercial endorsements? His annual 2 month vacation on Maui while every other manager in MLB is preparing for Spring Training? When Mike Myers says the team suffers from lack of concentration on the field and bad fundamentals...who is to blame? Do you think maybe not a lot happens during Yankees Spring Trainings? Giambi's half-joking reference to looking forward to his first ST to see Jeter and company practice the "flip" play resonates now with an ominous ring of truth.

66 I'm a stat guy and it is true.

See how well that works!

What would you call an average of 10 individuals' testimony? :)

Me, I was restricting my use of "stats" to baseball. And me, I'm very comfortable in trusting what the SABR community says and what's presented on BP.com - much more so than other source for baseball "truths".

As for the BCS - it's not much different. Knowledgeable individuals make a more informed deicision than one alone. As the poll is codified and weighted, it does an even better job. But the emotions of individuals gets in the way of accepting the most optimized answer.

69 huh? be fair.

68 Oh, and by the way, since folks seems to forget:

The Yankees lost the ALDS in five games in 1997. Jorge Posada - of .250 .359 .410 (188 AB) that year - got 2 AB's that series. Joe Girardi - of .264 .311 .334 (398 AB) - got 15 AB (2 hits).

Torre has always been this way. It's just gotten that much worse with age and more decisions of the same type.

68 How many CF had 5 consecutive .900 OPS seasons? Bernie was a force. Of course, I though Beltran would have been perfect to replace that production, but Cash had other ideas.

It seems to me that getting a good offensive performer at CF, SS ,2B, and C(maybe 3 of 4) and then filling in the rest of the team would make a lot of sense. Signing corner OF and corner IF to huge contracts may not be as helpful.

I am not familiar with RATE.
Is it a purely Defensive stat?
Do you know where there is a definition/formula?

Frankly, I don't really trust D stats. I thought we agreed that because parts of D is based on value judgement (is that an error? was that in his 'range', was the sun really in his eyes, etc) that they were subjective.

How many times have you seen what YOU thought was an error called a hit? And vice versa?

Positioning, routes to balls, jump, and other factors all effect the outcome of a defensive play. These things are very hard to gage and qualify/quantify, but they are an important part of a poor/great fielder.

From my (subjective, man-love, LSD corrupted)memory, Mantle, before his legs gave way, was an above average fielder. He had great instincts, and speed when he was young. He had a GREAT arm. He may not have been great, or even excellent, but I can't see him as below average.

We should have an indepth discussion here about STATS. I believe many people misunderstand their use and how to evaluate them. Numbers don't lie. 150 hits / 500 ABs is a .300 BA. It's hard to debate IF 150/500 = 0.0300. But what does a particular .300 mean? How does any one person evaluate it? What other stats are used in the evaluation? what factors went into the 150 hits? What other factors could 'skew' the numbers?

Someone posted that Minky was a below average fielder... based on stats. Many of you saw how he played for us. Did you think he was below average? Are the eyes always wrong?

Stats are not perfect. The math (by nature) is, but the conclusions we reach are highly subjective. And if you add human factor into deciding the numbers themselves, it can bcome highly flawed.

Quick example. ARod has played well based or errors. I saw the first 3 errors he made. All were questionable, one was outrageous. My feeling is he has played better D then his numbers say. Small example.

I have a great 'Car salesman' story where a guy tried to prove to me that borrowing $20k at 9% would ultimately make me more money then paying $20k cash for the car. His numbers worked. I ran then different ways, changed the numbers slightly, evaluated the answers, and still came up with a profit.

I actually lost sleep over this, knowing it couldn't be true. But the numbers kept proving it was. Until I found the clitch. Car salesmen can be pure scum.

73 You're exactly right. The 1996-01 Yankees had exactly the right players to build a dynasty around - CF, C, SS that could hit a bunch. The greatest closer of all-time certainly helped too.

The only thing missing was an ace of the pitching staff. But they did very well to get a bunch of #2's and #3's.

A-Rod has a RATE of 100. That seems about right.

Mghkdfg didn't get to very many balls. All he did was scoop them. RATE takes the former into account. And shows he's been below average for years.

No one said Mantle was bellow average in his early days - he was about average and with a killer arm. But Mays got to many more balls. And the stats (RATE included) show so.

People are neglecting to realize that some things are easier to see with our eyes. And so the mind weights them more heavily. Scooped throws are easy to see - when they're made and when they're missed. Balls that get by are much harder to judge with our eyes. But the stats show how many "good" 1B's and CF's get to. Given the limited space of the diamond, the stats work better.

44"The point is: Just because some geezer said DiMaggio was fantastic, I shouldn't believe it."

What if that geezer was a former outfielder? Who saw the majority of games, in person, Joe D played?

What if the 'official' scorer was a geezer? Who didn't like the Yanks? Or Italians? Or wanted Mantle to have better stats?

What makes you think that the people who decide errors, rate, range, etc are not flawed? Or prejudice?

If you don't know how RATE is compiled, how can you have absolute faith in it?

44 One of the problems, perhaps, comes from the fact that said geezer uses the word "fantastic" and leaves it at that.

What's probably meant is something like "spectacular" which may or may not be the same thing as "excellent."

In other words, a fan's memory may be selective when reminiscing about generalities (good/bad), but if you ask the fan a direct question, such as, "How was Jeter at going to his left?" the fan, if he's honest and moderately perceptive, should be forced to concede that he was pretty lousy at it.

You may not be able to rely on memory to give the one-word assessment of a player's career, but you may be able to rely on that memory to tease out the overall picture by asking not about the spectacular, memorable plays, but asking rather about the routine ones.

Does this make sense?

74, 77 i agree. that's why i think we need 41.

76 yeah, but does it take "everything" into account? the speed of the ball? the spin? the position of the sun? the hop it took off the grass? the length of the grass? the baserunner blocking him? these things all have to be accounted for, and i very much doubt any existing defensive stat is all-encompassing.

77 Putouts and assists aren't decided. And both tell you alot - namely how many outs did the fielder make.

I'm pretty sure RATE is some derivative of range factor, since they closely match.

RF = (Putouts + Assists) x 9 divided by Defensive Innings Played

RATE just normalizes everything to a 100 scale to help it make more sense. Think about the role of the "9" in that equation above.

So, I put faith in RATE because the numbers aren't decided. And because, I believe the SABR/BP wants to get it right in a big picture way - much more so than yours or my opinion on Mghgdkj.

As for 1B - Mghkdghj may be the greatest scooper in all of baseball history and the current stats don't capture that, probably becuase it's so hard to measure. But that complaint only applies to 1B, and maybe C to a degree. But he still sucks at getting to balls hit at him - indeed, he's below average.

76 another question: while you don't trust some old geezer's eyes, do you trust your own eyes over the current defensive stats?

76 "People are neglecting to realize that some things are easier to see with our eyes."

I'm not disagreeing with your basic argument, that statistics allow analysis to step outside individual limitations of observation and biases. All I'm suggesting is that some statistics in baseball--mainly defensive stats--have a subjective quality built in, and we should caution against 'stat-triumphaism' as a result.

This is also true for stats based on recording specific sorts of events that only began to be recorded in the (relatively) recent past. Sometimes you can work backwards and try to find or 'recreate' the missing data, but sometimes it's impossible (we'll never know how many HRs Ruth gained from balles bouncing over the wall or lost because the ball landed foul after crossing the fence).

In some cases, we have to rely by and large on personal testimony--for example, with Nego League players whose exploits were not systematically recorded.

Ultimately, I was not challenging the assertion that DiMaggio or Mantle were overrated in terms of defense. I was, rather, raising a philosophical or methodological question.

79 Charting balls is ZR - Stats Inc owns the method and the formula. That's their attempt to get it right - by placing their own scorers at the ballpark.

73, 75

Are you saying that average RF, LF, 1B + superior CF, CC, SS is better than average CF, CC, SS and superior RF, LF, 1B?

Why isn't it more a question of finding 3-4 superior VORP (and defensive) players, regardless of their positions?

45 Well, if you're going to talk about cameras, I'll go you one better: they should rely far more both on the overhead camera (above and behind home plate) and that catcher-cam thing they sometimes use.

They use these other angles very unimaginatively. Randomly they'll use another angle, but it's only for one pitch and then it's back to the centerfield camera.

Why not use these other vantage points for the duration of an AB? It gives you a fresh perspective on the game. It's really cool to actually see an AB from the perspective of the hitter rather than that of the pitcher.

Also, they must really hate outfielders because that's the part of the game that's most poorly captured by the cameras. It's so hard to evaluate outfielders on TV because they rarely let you see the kind of jump the outfielders get and the routes they take.

If they got some imaginative directors with a feel for baseball, the games could be much, much more compelling.

since "mientkiewicz" is literally on the screen when you type a post, i wonder if anyone could try to actually spell his name. it isn't that tough.

80 Aha! But I thought the big knock on RF was that it was team dependent (sort alike RBI but not exactly). Since there are a finite number of outs, if one player has a high RF (that is, he makes more putouts than normal), he essentially takes away outs from other players.

81 No. I watch as many games as I can, and my eyes can only be looking at one place at a time. Besides, I've seen some pretty crazy things with my eyes thanks to psychotropics.

85 if i could have it my way, every single pitch would be shown from the true cf camera. i HATE it when they show it from behind, because it gives a worse angle. if they want to show a replay from behind, that's nice. YES shows a pitch from behind the catcher sometimes, and you can't even see the strike zone. it's infuriating. i do wish they'd show replays from the overhead camera more often, but we should also keep in mind that since that camera is from behind the plate it is not 100% accurate. i think all new stadiums should be built with cameras straight over home plate, 1b, and 3b, so that we can clearly see if a pitch crossed the plate and if a ball hit down the line was fair or foul. since i also think all new stadiums should be built with retractable roofs, it wouldn't be a problem.

84 I agree with you. I think the CV on this is:
1) 2nd, SS and CF are important defensive positions then the others.
2) Because of (1), these positions are usually manned by 'lesser' offensive players. Therefore, if you HAVE better offensive players at these positions, all things being equal, it's easier to find good offensive players at the other positions.

But I agree in that it's about putting the 9 best offensive and defensive players on the field. The positions is not as critical.

However, great D at SS and terrible D at 1st is better then terrible D at SS and great D at 1st.

RF is really not very accurate and is a misleading stat. what does it really mean? if a lot of balls are hit at you you have a better range? you can say it should all even out, just like bad calls should all even out, but how can it possibly?

84 I'd say it's easier to find the former than the latter. Or: once you have above average CF, C, and SS, you can really help yourself to find an above average RF (O'Neill) and 1B (Tino). The LF in those days always seemed to be below average - that's the position they carried - and it's just as silly given how easy it is to find a decent one.

86 Who?

87 I think that's true to some extent. But two ways around that: Look at the player over many years (thus different teams) and consider their position.

91 Look at the formula - it simply calculates how many outs a fielder has their hand in. Across players and years, you get a good sense of how "good" their fielding is, and much better than someone saying: "Mfhsjfkg is a great fielder". If you care about the 1B fielding balls, he's not.

Why rely totally on cameras? If you shoot laser light straight up from the edges of home plate, you can 100% accurately determine if a ball is over the plate. If any part of the 'beam' is interrupted, then the ball crossed, or nipped, the plate somewhere.

It's SIMPLE SIMPLE technology and 100% accurate. That takes care of inside/outside. Hi/low could be called by umps as today, or with cameras. However, if umps could totally ignor In/Out and just concentrate on Hi/Low, they might do a better job at that aspect of it.

And I think the more aggregious calls are made inside/outside.

94 what about checked swings when the bat or an elbow touches the strike zone? is there a way for it to ignore that and only sense the ball?

87 92 93 And, RF gets clobbered by teams with high K pitching staffs. This could be easily fixed by calcualting RF/outs in play, and then normalizing the result (maybe RATE does that?).

90

Thanks for that.

I guess another way to look at it, in order to reflect the premium value of certain positions, is that a really, really good 1B, RF, LF is as valuable as a good SS, C, CF.

It's all relative. I'm just reluctant to say that the "formula" is great CF, SS, C + others (73 and 75).

65 What's up with that, JL? I'm "people here" and I certainly don't know better, fwiw.

Hang in there.

Speaking of Tino and O'Neill: Look at those acquisitions. When's the last time Cashman traded some talent to get in a younger player with upside? That's as big a problem now than anything else - he filled the team with with short-term solutions (Abreu, Damon) and took too long to do so.

What they Yanks have really needed the last few years are exactly those sorts of trades for CF, 1B, BUC, and now RF.

93 your logic is very inconsistent. at one point you say everything should be stats and you shouldn't trust people. at another you say you just trust the people at BP to be right. then you're talking about getting a good "sense." and this isn't just a poor choice of words, it's true. all you can get from that stat is a sense. it doesn't really mean anything at all. if you are playing ss behind cm wang, you are going to have a ton of assists and putouts. multiply that by 9 and divide it by innings pitched, it's still gonna be a lot higher than other shortstops playing behind a fly ball pitcher. or the same ss playing behind a fly ball pitcher. and it just can't all even out in a way that makes the stat accurate. all it is is a sense.

74 I'm not sure what you mean. Plenty of people have analyzed stats. In baseball, the two most important quantities are runs and wins. People have done correlation studies of a given stat with the number of runs scored/prevented, and then these numbers with a wins/losses. The greater the correlation, the better the stat. You can quantify it completely. There's no opinion in there. That's what gives us EqA as a better stat than OPS (even though OPS is pretty good). EqA correlates better.

94 One of the 'problems' with balls and strikes in baseball is that, for whatever reason, the Abner Doubleday and all the other mythical founding fathers decided that the strike zone should be relative to the batter. In cricket, by contrast, you try to knock over the wicket--a fixed bundle of sticks whose size is uniform no matter who s batting.

I wonder what would happen if the league decided to go to a fixed strike zone. It would be much easier to umpire electronically, as sensors could easily set up to determine if the ball crossed through teh zone. But obviously some players would be hurt (and helped). It would eliminate (possibly) some of the more extreme stances, like Rickey Henderson's, which might be a shame.

99 paul o'neill was 30 when he started playing for the yanks. tino was 29. abreu and damon were both 32. does it make that big a difference? personally i never wanted damon, but i thought abreu was a good pickup.

102 i think that the sensors could adjust to a batter's height and accordingly measure strikes between his knees and chest.

96 I'm not sure. But again, that's where different teams tell us everything we need to know. And Mghdkj hasn't changed over his last four teams.

97 I don't think I was saying it's a formula. You can't plan for a great SS, C, and CF, and certainly not all at the same time. Just that that accounts for the success of the Yankee dynasty as welll as anything else. And once you have that, it's easier to fill in the pieces of an above average RF, 1B, and even 3B.

Keep in mind that the Yankee dynasty had a top 3 SS, C, and CF - that's unheard of.

94 RFID technology would work much better. They replace the balls enough to prevent damage. And you can always track it. That's how they did the highlighted puck on NHL broadcasts a few years back.

104 But remember, the zone is technically the height of the ball when it crosses home reltive to the batter when he swings (normally). That's prety tough to adjust for with sensors, though I guess they could produce a massive database of all hitters' regular swings, and teh relative heights of their strike zones.

102 I think they should just stick a lawn chair back there and if it hits any part of the chair, that's a strike. It worked for me when I was 12.

76 No doubt, Jim, some things are easier to see with the eyes, but the eyes can be trained to see things it formerly might not have.

It's just like with music or art. A novice can listen to a symphony and hear only the primary themes or melodies, but if you listen a lot, to a lot of different kinds of music, and talk to people with keener discernment, you can learn to hear the entire orchestra simultaneously, or to focus your attention on the woodwinds, or whatever.

Same with visual art. You can train your eye to see nuance.

Comparison helps.

With baseball, you can watch a bunch of different teams and you start to get a sense of what's possible and what's not.

It's hard to watch Reyes play short, for instance, and not be struck by Jeter's limited range by contrast. You get used to seeing Reyes make these plays and then you see the ball get by Jeter and you realize that a better fielder would have had the ball.

So what you're saying makes sense if your premise is that stats are better than the naked eye of what you seem to think is the casual observer, but if that naked eye belongs to the discerning or sensitive observer, then I'm not sure your premise holds.

107 not the legs

Isn't the "number of runs scored/prevented" highly circumstancial? How good is the offense? The defense? ERs are dependent on errors, which is subjective. Cheap hits... loud outs. How about bad calls? How many runs have the Yanks lost on bad calls.

There are many factors that go into the total game. Stats do the best they can, and are probably accurate at times (but which times) and probably representative a lot. But they are far from absolute.

And of course, some are much better then others. As I said, BA is perfect. One number divided by another. But what makes a hit is sometimes subjective. A guy who posted a .280 BA could be better then a guy who posed a .300 BA, but with worse luck.

And is not that that stat is flawed, as much as the 'conclusion' any one person reaches.

100 Look at the stat over teams/years if you think one pitcher has a huge effect. The information is in the data. It's not in your eyes or your head. The stats give you the best chance.

103 I would have rather had Damon in 2004, yes. Abreu too.

105 On RFID, you could also put the chip in gloves, in bases, in shoes. You would "know" if the ball reached the glove before the for hit the base. It wouldn't work for tag plays, but then the cameras, and slo-mo, already do a good job to help the humans.

107 Finally.... a man of science.

108 You're right, but then one set of eyes has to watch all those games. The first folks to "see" Jeter's limitations were probably of this type. The stats just help the rest of us catch up.

Meanwhile, why I restrict discussion of the stats to baseball is because the discrete events make them that much easier to track. Baseketball and hockey seems near impossible to use the same level of analysis. There's more hope for football, but even then I'm skeptical.

Cricket I'm fascinated by. I'd love to play. Great point 102 by the way.

107 :)

By the by... I took a look at the 2008 FA 1st basemen. I'm not really up on many non-Yankee players, but I didn't see a lot I liked. Is there an answer to our 1B prblem there?

Someone help me catch up here. Cashman should be canned because he wants to use a lawn chair and a camera in CF to calculate RATE, instead of the SABR-charted batted ball data from the 50s?

I just scanned through the comments, so I might have missed some of the details, but I think I have the gist of the discussion. Right?

;)

111 yeah i pretty much agree with you, but it is just a sense. the stats are far from perfect, but i think with some of this rfid stuff they could be, because instead of trying to record all the information through various stats, we would record all the information first and make stats later.

113 My buddy's girlfriend (now wife) is Australian and ahuge sports fan, espcially cricket. She of course got really angry any time I compared cricket and baseball ("that's the problem with Americans..."), but the two sports are very interesting to compare (and contrast).

Cricket is geared so much more to offense, while baseball still (despite the current era) is geared toward defense. That's why cricket matches take so long; the shorter matches are intriguing: they put a limit on the number of "pitches" (using BB terminology) thrown, rather than on the number of outs. This changes the game significantly because it changes the limiting factor--this in turn convinced some teams to change the way they played. So, teams would play their defenders back, conceding single runs in order to prevent fours (when the ball rolls over the line). This then compelled the league to mandate where the team could station defenders--they were trying to legislate a certain style or aesthetic of play while at the same time changing the conditions that led to the evolution ofthat aesthetic. Fascinating really.

When I pointed this out to my friend's wife, she told me that I just didn't understand cricket.

115 Free agency is part of the problem - not the solution. See again: Martinez, Tino and O'Neill, Paul.

For those still looking for an explanation, RATE is a Baseball Prospectus stat that indexes the number of runs above or below average a fielder is per 100 games. A fielder with a RATE of 110 is +10 runs per 100 games, a fielder with a RATE of 90 is -10 runs per 100 games.

RATE is based on BP's Fielding Runs Above Average stat, which is NOT based on play-by-play data because said data only goes back so far (1959 in the BP database, more or less in tune with how far back Retrosheet has gotten). The function of FRAA is as a fielding stat which can cope with the whole of baseball history going back to the 19th century. Basically, the underlying formulas are based on the "deltas" of plays made versus expected plays made (putouts, assists, DPs) based on the number of runners a team allows on base, the lefty/righty balance of the pitching staff, the implied groundball/flyball ratios (not actual ratios since again, those are PBP-based) based on team pitching statistics, etc. I believe the basic formulas were published in Baseball Prospectus 2002 or BP 2003, which is something of a pain for anyone without a copy.

FRAA and RATE do a decent job as an all-time stat, but they have their vulnerabilities. The discretionary choices as to which of two adjacent fielders handles certain balls (DPs and pop flies, particularly) can really send the stats out of whack. To use Jeter as an example, the change between having Ventura and Soriano as neighbors in the years prior to 2004 and A-Rod and Cairo/Cano since is somewhat responsible for the drastic turnaround in his stats and is probably overstated.

As somebody who uses the BP fielding stats in industrial-sized doses, the best I can counsel those of you who wish to use it is not to get too hung up on one-year fluctuations or small sample sizes. You'll get a truer sense of a fielder's ability by using a moving three-year average of the FRAA or Rate stats.

Thus, instead of reading Jeter's FRAA since 2003 as -23, -5, 13, 9, 9, read as -17, -5, 6, 10, and [too small a sample size to evaluate]. His RATE (FRAA per 9 innings plus 100) based on those numbers would be (and I'm using a spreadsheet here) 91, 97, 103, 108 and N/A instead of 80, 96, 108, 106, 113.

Sorry if that sounds overly complicated. It is, but with a bit of massaging, it's a reasonable estimate that works across a large span of time AND is less vulnerable to season-to-season swings than the way it's presented on-site.

110 If you take a large enough sample of teams, over a large enough number of years it tends to even out. That's why you have uncertanties and things like standard deviations. Look its not perfect, but its a lot better than me watching most of the games and saying that the Cairo guy sure is a great situational hitter!

116 You missed the part where Jim Dean complained Cashman should have gotten more than an old lawn chair and a camera for RJ.

118 Excellent, thanks. Maybe I found a new past time for the summer?

116 No, that's basically right, Shaun.

Except you forgot about the part where Cashman argued that if those old Yankee teams had had Lofton in center instead of the Mick, they'd have been even better!

Those teams won despite the Mick, regardless of what some geezer's lying-eyes-based selective memory might say.

Good job catching up, btw.

116 I'm just touched that my only two posts (about Mickey Mantle's RATE and lawn chairs) out of 100 or so influenced your impression of this thread.

And you forgot about Cricket. Cashman should be fired because he can't find a good back up bowler (BUB).

120 Thank you for this!

120 Excellent, thanks. It's too bad the formulas and databases of go betweens aren't published - they'd increase adoption and dissemination, and maybe even improvement.

What explains the apparent correlation of RF to RATE then? And you know what that number is? RF is at least very easy to get our heads around.

Oh, and Mantle and Dimaggio look average in CF. Any reason we shouldn't "trust" that?

121 Or that Cairo is a good 1B.

122 Nice. But he should have gotten some wickets too.

120 Thanks, Jay, I'm sure we all appreciate the big guns coming in now and again to set the little guys straight ;-)

Anyway, what are your thoughts on the historical data for RATE? Often I see mentioned in BP literature that they do a "nose test" to determine the validity of a stat. If all the "geezers" are saying that DiMaggio was a graceful CF with amazing abilities, and RATE says he's simply average, does that mean the Geezers are wrong or that the input data for the 1950s may be flawed?

120 Thanks for the clarification on RATE, Jay.

122 124 125

Ah ha! So, Cashman should have asked for a cricket bowler and an rfid tag from the Diamondbacks, instead of the camera and the lawn chair. Then he could have argued that the Yanks of the 50s should have used that bowler and Lofton (with the rfid tag implanted in their shared glove) in a platoon in CF (Casey loved platoons) instead of the Mick, leading to even more Serious wins.

OK. All caught up. Thanks all!

I just wish I had something to contribute now.

BTW, if you haven't voted for Jorge 25 times already, what are you waiting for?!?!?!

http://tinyurl.com/3xy7r6

Posada starting the All-Star game would be awesome.

130 Done, but only if I can write in Cairo at 1B 25 times. I also voted for Milton Bradley, Darin Erstad, Scot Podsednik, David Eckstein, and Corey Hart (Because he wears his sunglasses at night.)

Wow, the discussion here is so so so much more fun when focused on things besides Cashman, Torre, and the probability of an aging veteran retruning market value in a dump trade. People show some humor, takes the mind of the total suckatude that is the Yankees, and, wait, what's that, a new topic too? Daaaaaamn...

124 Not to say if they'd only had a decent BUC. That damn Yogi catching 150 games, what was he thinking??

132 Well, Zack, if it's amusement you want...

Polar bear walks into a bar.

Bartender says, "Hey buddy, what'll you have?"

Polar bear says "I'll have a gin............................................................................................................................................................................................................and tonic.

Bartender says, "Sure, buddy, but what's with the big pause?"

132 Well, Zack, if it's amusement you want...

Polar bear walks into a bar.

Bartender says, "Hey buddy, what'll you have?"

Polar bear says "I'll have a gin...

Oh no, I killed the joke!

Damned computers.

Quick, let's get to 161 comments and fix weeping's mess! What's a great way to get debate.

Hmm... Wil Nieves belongs on the Yankees!

Bring back Bernie!

How about this: Last night Jerry Marron, Reds' manager, said that David Weathers was a "mix of Rollie Fingers and Goose Gossage." Huh? David Weathers?

136 Well there was that stupid article that basically said the Yanks haven't won b/c they don't have enough guys like David Weathers. Yeah.

133 I don't get it :)

Wait....

Something about a bear and a pause...

Or is that..? Yes! Not pause...

Paws!! See, its funny because they sound the same! :)

I do think that the delivery really made that joke though brunnhilde!

133 TinyPAUSE.com is your friend.
;-)

I think I killed what was, up until my first post, a lively discussion. I better go away now and allow it to return. Proceed...

136 Did David Weathers grow a funky mustache or something?

Big PAWS!

Big PAUSE!

Eh?

Polar bear says "I'll have a gin..." and then just stares at the bartender for like a minute (works better orally, you see) until finally completing his order, "...and tonic."

Big PAWS!

Anyway, the poor polar bears are drowning these days because the ice is melting.

So sad.

144 Do not, under any circumstances, do a Google Image Search for David Weathers.

Scary.

145
Science Joke Alert:

But its okay, the white bears mix good with water, because they're polar.

/science joke.

68 "Does any other team of the free agent era feature three HOF in their prime?"

Do the Mariners of 1996-98 count? Griffey, Martinez, a young Alex Rodriguez in the lineup, and Unit in the rotation...

68 148 The 1995-1998 (or so) Cleveland Indians are interesting, depending on whether Thome (possibly) and/or Vizquel (reputation but underserving?) and/or Belle (career too short, no one likes him) join Manny. They also had Sandy Alomar.

I'm not saying they all should be HOF, but they might be 'name brand' enough to get in, at least some of them.

148 Somehow I missed this, as it makes for good post filler. I think I read on BP that on average there are 50 HOFers playing at any given time in the majors. That works out to 1.67/team, so surely there are examples of teams with 3 in the past 25 years. How about:

2006 Astros: Clemens, Biggio, Oswalt?
2005 Mets: Piazza, Glavine, Pedro? (And Minky and Cairo! How'd they ever lose!)
2004 Red Sox: Manny, Papi, Pedro, 38pitches.com?

147 Good thing I'm a humanities person because I so don't get that joke.

Late 90's Braves: Maddux, Glavine & Smoltz (also Chipper Jones, Andruw Jones & Fred McGriff ?)

152 2002 Braves: replace McGriff with Sheffield

to add a little more to the discussion on appreciating and understanding, here's an excerpt from a Tom Tippett article explaining how Diamond Mind uses play-by-play data to derive their defensive metrics, as well as examining overall questions and difficulties inherent in measuring defense AND examining the methodology used in deriving a range of existing defensive metrics. A very good read.

http://www.diamond-mind.com/articles/defeval.htm

"For a few years now, I've wanted to write a little piece about how difficult it is to judge defensive ability, or any baseball skill for that matter, just by watching a lot of games. Then I found an essay by Bill James in his 1977 Baseball Abstract (a self-published book that predated his debut in bookstores by about five years) that says it far, far better than I ever could.

Here are a few excerpts from this wonderful essay, starting with a comment on how differently most people tend to approach the assessment of hitters and fielders:

"While we might not all be able to agree who the greatest-hitting first baseman ever was, the record books will provide us with a reasonably brief list to choose from: Gehrig, Anson, Foxx, Sisler. That's about it. Nobody's going to argue that it was Joe Judge or Moose Skowron, because the record books simply will not permit it . . .

Fielding statistics provide no such limited clarity. Talk about the greatest fielding shortstops ever . . . and the basic argument for everybody is 'One time he made a play where...'

Suppose we turn that same argument back to hitting. Now Moose Skowron hit some baseballs a long way, but nobody is going to say that he was the greatest hitting first baseman ever because 'One time I saw him hit a baseball so far that..." It is understood, about hitters, that the important question is not how spectacularly but how often. Brooks Robinson is known as a great fielding third baseman not because of the number of plays that he makes, but because he looks so good making them. Nobody talks anymore about what a great hitter Jim Northrup was, although to tell you the truth I never saw anybody who looked better at the plate. It is understood that, notwithstanding appearances, he wasn't an especially good hitter. Hitters are judged on results; fielders, on form."

And he talks about the difficulty of trying to judge effectiveness simply by watching:

"One absolutely cannot tell, by watching, the difference between a .300 hitter and a .275 hitter. The difference is one hit every two weeks. It might be that a reporter, seeing every game the team plays, could sense the difference over the course of the year if no records were kept, but I doubt it . . . the difference between a good hitter and an average hitter is simply not visible."

"a fielder's visible fielding range, which is his ability to move to the ball after it is hit, is vastly less important than his invisible fielding range, which is a matter of adjusting his position a step or two before the ball is hit."

In that essay, Bill went on to propose a scoring system that accomplishes essentially what STATS Inc. is doing now -- recording the location of every batted ball so that we could build a record of fielding performances similar to the statistical records that we use to judge batting and pitching performances.

I'm not saying that it doesn't matter whether you watch games or not. I'm just saying that I agree with Bill that it's very difficulty to rate players solely by watching games. We also need useful measures of what they accomplished."

152 2000 Baltimore Orioles: Cal Ripken, Jr, Sydney Ponson, and Karim Garcia?

1982 Brewers had four HOF'ers on the roster: Robin Yount, Paul Molitor, Rollie Fingers and Don Sutton (only pitched in 7 games)

155 LOL. You got me with that one...

They also had Moose and Scott Erickson...

127 While RF or Range Factor does measure the raw number of plays per game a fielder makes, it does not account for groundball/flyball distribution of balls in play, the left/right orientation of a pitching staff, or the staff's strikeout rate, while FRAA and Rate do. All of these can greatly affect the number of balls in play available to a given fielder over the course of a season.

As for DiMaggio and Mantle, there's no denying their physical grace in the field, but let's remember that the accounts of their fielding prowess came from sportswriters who were 1) considerably more NY-centric than today; 2) at least somewhat more beholden to hyperbole than writers in this day and age; and 3) less prone to being second-guessed by TV media and viewers. Furthermore, said writers' creation of and adherence to the conventional wisdom is exactly what statheads have been battling against for years. As James said, there's no way to visually distinguish the subtle difference between a .300 and a .275 hitter, and the same is true for fielding.

Once you put their teams' numbers through the ringer, Joe D. winds up as slightly above average while the Mick is well below; the latter probably should have been moved to a corner earlier in his career. Now, the odd configuration of Yankee Stadium may have something to do with these numbers, though that's likely already reflected in the breakdowns that go into computing FRAA.

The bottom line is that nobody's wrong here, we just have more information at our disposal. It's perfectly correct to say that in his day, DiMaggio's grace afield transfixed observers and he was hailed as the best in the land, but the data attached to those years doesn't entirely support the level of hyperbole.

Grrr. Damned if I'll ever figure out how to properly link back to previous comments here. What's the proper coding?

open and close straight brackets, jay, with the number in between

How many posts does it take to get rid of the wide format?

161 I guess I answered my own question...

158 Excellent again!

Any chance you know the correlation between RF and RATE. I'm guessing somewhere between .7 and .8.

148 Excellent! If management had found a way to keep them together, they may have just won a title or two. Problem was: Keeping them together would have meant little cash to support the rest of the team. Problem was: They didn't all come up together so it was only those three years. And Alex had yet to mature as a hitter.

150 My qualification was: in their prime

The 2004 Sox had Manny, Pedro, Shrill at the end of their primes. If everyone was a few years younger, you'd have to think they would have won another title or two (shudder).

154 Excellent!

156 That 1982 Brewers team is a great example. Yount and Moliter - great combination. Except they only controlled Fingers for three years. If only they had a few decent pitchers to front the staff. They were AL Champs with only one above average pitcher (Pete Vuckovich) and barely at that.

Recall the hypothesis: The 1996-2001 Yankees benefitted from controlling a few exceptional players (Jeter, Mo, Jorge, Bernie) in their prime. It's hard for me to ascribe credit to the manager for falling into that windfall. And Pettitte was a consistent above average pitcher. Clueless Joe may have just been more lucky than anything else to have a great team handed to him.

Meanwhile, looking back, the only thing that may have kept that 1995 team from going further was playing Tony Fernandez (season 76 OPS+ and 5 for 21 in the ALDS) instead of Jeter and not having found Mo's setup role.

163 That and Showalter's short memory regarding Wetteland. Though he got torched in gm 4, he did pitch 3 scoreless innings in game 2.

Days like today it's not fun to look back and see the parade of Yankee SS of my youth

Alvaro Espinoza
Stanky
Spike Owen
Gallego
Tony Ferandez

Jeter's certainly changed expectations.

163 No idea on the correlation between RF and RATE, but I wouldn't be surprised if it's that high. Nonetheless, that alone doesn't make it a particularly useful tool for comparing players on different teams given the variables I cited above. Two players can have similar range factors yet wind up with divergent RATEs.

Jeter's been a long time coming, SS for the yanks has been a long parade of replacement level crapulence in my lifetime:

Dent
Smalley
Meacham
Tolleson
Santana
Espinoza
Stankiewicz
Owen
Gallego
Fernandez
Jeter

Sheesh ...

163 Or he could have stuck with Mo longer in Game 5. I always forget that's the series the greatest closer of all-time was discovered.

166 Thanks but then those variables would drive the correlation lower - i.e., If the RATEs are different but the RFs are the same, they'll be less correlated.

Still, I'm very cool with RATE - easy to understand and reliable. I think I'm more asking for the folks here who are uncomfortable using it.

The problem with RF, I think - as seen on BR site - every position is different. He's done a good job there to show the league average for the position, but it's not easy to follow.

As for RATE - what would you say about a guy like Cano. 2.5 years in the league. He seems to have improved dramatically since 2005. Too early to say?

Despite my earlier retirement, I've been convinced to try again. I feel compelled to add one more post here - a very, very long one.

I don't actually claim that eyewitness testimony is more accurate than fielding statistics, only that I don't trust those statistics. (I may have fallen into that trap with Joe and Mick, but generally I avoid it.) Simply having a number - even one computed by experts!!! - isn't necessarily better than not having a number. I've worked enough with statistics, and with overblown claims for them, that I always approach them skeptically. From what Jay's said, there's still quite a bit of subjectivity involved in calculating RATE - especially historically,
when you have to deal with things like implied ratios. At that point, I start losing a lot of confidence.

Another problem with fielding stats is that there's no external criterion to guide us. An earlier poster talked about the correlation of offensive stats to runs scored, which provides validity for
the stat. With fielding stats, we can correlate them to...well, to nothing. There is no concrete measure of runs prevented as there is for runs scored.

Finally, RATE actually tells me next to nothing about a player's fielding; it just gives one big number. There are no components to it that would help me understand it. I've done a lot of IQ testing, and the single least interesting bit of information is the overall IQ, the one that says that a person with a 100 IQ is "smarter" than a person with a 90 IQ. I barely pay attention to it - it's the
components that make up that score that provide the useful information. I find BA/SA/OBP better than OPS, and if EQA is "more accurate," so what? The more manipulation that goes into the stats, the wider the confidence range is going to be (not that baseball analysts pay any
attention to confidence ranges, or significant differences, or anything like that.) Does it matter if a "more accurate" measure yields an IQ of 103 rather than 104?

If I think a stat is limited or subjective or overly mysterious (as in the proprietary stats), then I may look at it, but I'm going to take it with a shaker or two of salt. I'm not going to take it as any hard-and-fast assessment of a player's fielding, because I don't really know what it means, or how valid and reliable its findings are.

I do trust my own judgment, my own eyes, but mostly after I've seen someone enough to have
confidence in that judgment. With other players, I basically don't make a judgment. I accept that there's no measure that I find terribly useful. The desire to measure something doesn't mean that any attempt to measure it is helpful. So is Ozzie Smith the best fielding shortstop of all time? I have no idea, and I'm not interested in arguments about it.

But...but...how can I just accept that there's no useful measurement? Hell, we do that with a lot of aspects of baseball. How can you distinguish an average outfielder's arm from a better-than-average one? How good is Tyler Clippard's curveball, and how much better or worse is it than Aaron Sele's? How good is Derek Jeter's range to his right, how strong is his arm? How good a defensive catcher is Jorge Posada, and how does he compare with Bengie Molina? Just how
vulnerable to outside breaking balls was Alfonso Soriano in 2003? Would it help if someone came up with A Number? I don't think so. I watch, and maybe look at stats, and I make my judgments cautiously.

I think it bugs me most when people throw around stats with only the vaguest understanding of what they are, what they mean, and how good they are. Stats are wonderful things when you understand them, but they can be extremely deceptive otherwise. They can also be used as bludgeons, as a substitute for thinking rather than one piece of a discussion. That's unfortunate.

167 LF is just as bad, since 1990:
Oscar Azocar
Hensley Meulens
Mel Hall
Dion James (I do not remember this guy at all)
Polonia
G. Williams
Raines (the bright spot)
Curtis (one good year, one bad)
Ledee
Knoblauch (wtf?)
White
Matsui

167 I was too young to remember Roy. But from the looks of things he's wasn't awful - like the guys that came afterward. I remember cheering for Meacham and all those other scrubs.

It was like having Cairo play SS for 12 years.

Wow.

170 "From what Jay's said, there's still quite a bit of subjectivity involved in calculating RATE - especially historically,
when you have to deal with things like implied ratios."

I see nowhere where he said anything along these lines.

"RATE actually tells me next to nothing about a player's fielding; it just gives one big number. There are no components to it that would help me understand it. "

I agree with you here. But whereas in other areas stats are used to advance an agenda, the BP guys have none but to descibe the game and it's history. I respect that. It's not like they're trying to make Teddy Ballgame look better than Joe D. The stats are as unbiased as they could ever be.
together?

As for all of your questions, I think there are very good answers that can be defined statistically. And those answers will be better than any one person's opinion.

171 At least half of those guys had above average offensive seasons though. Still, you're right. 2B was pretty bad for a while as well.

168 Looking at the box score, I thought Mo pitched longer than he did. Interesting that McDowell got the call before Wetteland did, and not only that, stayed in after the Yanks got the lead.

167 I still don't know what the Yanks were thinking signing Gallego to play SS. They knew from the beginning that he didn't have the arm to do it...

171 Dion James was the guy who killed a bird with a line drive @ Shea while a member of the Braves. He was pretty much a part time platoon lefty bat, high average, no power. I think he was to be platooned with Gerald Williams, but there was so much flotsam & jetsam floating through LF at the time, I may have been wrong. IIRC he was busted for pot after the Yanks cut him.

172 IIRC, Smalley was a good bat, passable glove type SS. Think his range was fairly limited, though my memories could be playing tricks on me.

173 Yeah I was impressed that this Dion James fellow came out of nowhere to have a 138 OPS+ in 1993 despite hitting only 7 homers in 343 AB.

Also his BR.com page is supported anonymously by someone who writes:

DEAD BIRD. GONNA PUT IT ON YOU FOR A DOUBLE.

I was 9 years old in 1993; is this a reference to something?

170 One more shot: All a defensive stat needs to tell you, for the most part, is how good is the fielder in converting outs. That's RF. FRAA and RATE make further adjustments.

If you want to see the strength of a guy's arm, see outfield assists. That told me earlier today that Mantle did have a fantastic gun.

Sure there will be other idiosyncracies at each position that's not fully captured (caught stealing for the catcher, picks by the 1B) but then there are sometimes other numbers to supplement them. Just cause there's one number, doesn't mean you have to be a slave to it.

But if I want to know wther Mghjfdk is a great 1B, I'm glad to know there's a decent answer and one that tells me he's far from great, no matter what a guy like PeteAbe says.

173 "Expected plays made" and "implied groundball/flyball ratios."

177 I don't think that's filling in numbers like you make it sound. I think that just means using the context of the other players on the team and the number of balls they got to. At least that's how I'd do it.

Help Jay? (And thanks for all of it!)

175 From Wikipedia; "James may be best remembered for an incident during a 1987 game when a fly ball off his bat hit a bird in flight, killing the bird. James, reaching second base on the hit, was credited with a double."

176 My understanding of Range Factor is that it's not a well-regarded stat at all. Bill James originally proposed it, used it for quite a few years, and then abandoned it. The problem is that there are far too many variables that affect it that may have nothing to do with a player's ability, particularly in the composition of the pitching staff - GB/FB. lefty-righty, strikeout v. contact, and so on.

Outfield assists will only tell if a player has a better arm than runners think. A player with a lot of assists probably has a very good arm, but a player with fewer assists may not have a weaker one. Also, it has such a narrow range that its ability to distinguish one from another is limited. And, finally, a lot of assists might show a great arm, but that was deliberately not my question. I asked about average v. above-average (not necessarily the best), and assists don't help with that at all.

178 Yes, I understand that, and I never said it was just filling in numbers. But there's no entirely objective way to measure that context. (How are you going to assess the fielders around him? By using their RATE?) That can only be derived by making subjective choices and decisions that I'm not privy to and hence can't judge. They may be entirely reasonable and unbiased choices - but there might be a lot of other options that are just as reasonable, just as unbiased.

176 actually, i'd say there are better metrics out there than outfield assists, metrics that measure an the effect that an outfielder's arm has on expected baserunner advancement. Assists are misleading, for example in that a poor throwing outfielder might accumulate more runner throw outs since he has more opportunities as more runners test his arm; conversely, a great throwing outfielder might not be challenged much at all, leading to lower assist totals.

there was a hardball times article i found and referenced during the great "damon throws like your weak sister" argument (short answer: he does!) a week or so ago, i see if i can dig it up again ...

are, here we are

http://www.hardballtimes.com/main/article/best-outfield-arms-of-2006/

176 Fielding batted balls is a much smaller part of a first baseman's job than receiving throws. So if RATE only measures batted balls, then it shouldn't be represented as showing how "good" Mientkiewicz is. It only shows how good he is at fielding batted balls, which isn't the same thing.

Any stat only tells what it tells. Someone asked me recently what IQ tests actually measure. The shortest and most accurate answer was, "They measure IQ." They don't measure how "smart" someone is. In just the same way, RATE measures RATE, and shouldn't be taken to measure "fielding."

Btw, a first baseman's ability to receive throws also affects the other infielders' ability to turn batted balls into outs.

180 RF is just easy to understand and follow. Yes, there are problems. I understand that which is why I appreciate RATE more and the people in the know (who care about getting these things right).

181 They could just use the numbers of balls hit to the surrounding fielders as input for the RATE (FRAA) equation. Nothing more magical.

182 Absolutely. My only point is: Just cause some stats have only one number doesn't mean you ignore everything else.

185 Except that your use of stats, in my experience, almost never involves much subtlety or nuance. You cite RATE to tell "how good" a fielder is.

RF is easy to understand and follow. It is also deceptive and measures noise much more than signal. RBIs and Wins are easy to understand and follow, too.

184 "Fielding batted balls is a much smaller part of a first baseman's job than receiving throws."

I can't agree with that. If they miss a ball to their right, it's likely a single. If they miss one to their left, it's likely a double.

A missed throw almost always means one base (it's a single), if the catcher is backing up. And it could have been a close play any ways.

Add those things up, and the missed grounders mean more damage.

"Any stat only tells what it tells. Someone asked me recently what IQ tests actually measure. The shortest and most accurate answer was, "They measure IQ." They don't measure how "smart" someone is. In just the same way, RATE measures RATE, and shouldn't be taken to measure "fielding.""

That borders on the nonsensical. If an IQ test includes measures of short-term memory, pattern completion, and math skills, then that's what's measured. RATE includes many facets of fielding and normalization based on what other players on the team do adn the rest of the league. That's good enough for me.

When if comes to judging fielding, you have two choices: a) Jeter is a (great) fielder because he makes (great) plays; or b) Jeter is a (slightly above average) fielder because (an approximate summation of all of his plays relative to the rest of the league) says so.

I'll take (b) every time, thanks, especially since I know all the guys at BP aren't Yankee fans and want to get the question as right as possible.

186 If the question is: Is X a great fielder, RATE will give a better answer than most other things. And unless RATE and ZR come up with different answers, then it's question answered. Two methods - same answer - I'm happy.

RF: I wouldn't advocate, unless someone is scared to use RATE or ZR (cause they're in a box). In that case, I'd still rather see something other than: "Jeter is a (great) fielder because he makes (great) plays".

183 It's an interesting effort, but no more than a first step - a hypothesis, really. I also looked at the article where he describes the nuts and bolts (http://tinyurl.com/oegze), and there are piles of subjective choices made - reasonable ones, but ones that could have been made any number of other reasonable ways.

Another problem that I have with baseball analyses is their complete disregard for things like statistical significance, variability, and error. For example: lots of talk about "small sample sizes" but no measure, ever, of how big a sample size is needed for a number to be reliable.

Similarly, there's no idea of how big a difference is a meaningful one. For instance: "Green can't throw anybody out (only one kill, when four were expected) and the runners know it." Well, how meaningful is the difference between four expected and one actual? How big s difference does there need to be to tell us something?

OK, enoughfor this one.

187 I am a Yankee fan, and I can give a much better answer than "slightly above average." But maybe that's because "how good," as a summation of all skills, is basically a boring question to ask in the first place.

And no, it's not at all nonsensical to say that IQ tests measure IQ. It's a way of saying that you can't take the one number to mean anything more than what it is. Even the components you mention - short-term memory, pattern completion, math skills - are made up of other components that might not be obvious. The test called "Arithmetic" on the IQ test I use measures caculation skills - but also listening comprehension, short term working memory, attention and concentration, and they're at least as important as calculation. It's only by looking at the details that you get meaningful information.

"Another problem that I have with baseball analyses is their complete disregard for things like statistical significance, variability, and error. For example: lots of talk about "small sample sizes" but no measure, ever, of how big a sample size is needed for a number to be reliable."

I agree with that - but it gets me more with the PECOTA projections. They should tell us how far off they are each year. Not constantly tweak the model with more parameters and assume every year is equivalent.

190 Problem is: As a Yankee fan, you probably can't tell me much about the other teams in the league. There'd be no way to know that Mgfjk is below average before he signed to be their defensive 1B. The stats tell us he was, and for at least the last three years.

191 But the problem with any composite stat is that the errors in the various components are summed in the total. That's why an IQ score is considered accurate within a range of, I don't know, 10 points each way. I much prefer to report a range - Average, Above Average, Borderline, whatever - because it lessens the false impression of precision.

192 Nope, I can't. But that doesn't make RATE better. It just means I know my own limitations. Do you know the limitations of RATE?

191 Also, when I say "error" I'm not talking about the discrepancy between a prediction and the final result. I'm talking about the error of measurement.

193 No doubt.

194 If it answers whether X is great, average, or poor - that's good enough for me.

I don't assume it resolves whether Melky or Coco is better. That's knowing the limitations.

195 Problem is, for the purpose of the metric, they're usually one and the same. You can minimize both types, but you're stuck with them either way.

170 Wb, JL.

I knew you'd be back!

;)

174 All I remember about Smalley is that he hit .262 one year (that's right, isn't it?) and he was a switch-hitter.

Whew, the fur is flying here...

177 "Expected plays made" isn't some subjective notion on what a fielder "should" do. It refers to things like the team's percentage of double-plays converted per runner on first base (via hit, walk, HBP, ROE) relative to the league average, or the number of expected putouts based on a team's lefty-righty distribution and infield-outfield split.

Likewise for "implied groudball/flyball ratio," which basically refers to the ratio of doubles, triples, and homers allowed per single, relative to league averages (which have changed as the game has evolved, of course).

Personally, as much as I use FRAA and RATE, i this day and age it's far more appropriate to focus on play-by-play derived defensive metrics. Right now I think the best publicly available defensive system is David Pinto's Probablistic Model of Range, which gives you not only an index of the number of plays made versus expectation but has nifty graphs that show where a player is better or worse than average on grounders and fly balls. There's also some good stuff to be said about Chris Dial's work at BTF, and Mitchel Lichtman's Ultimate Zone Rating, at least before it went proprietary. I haven't seen enough of what David Gassko has done with his defensive work at THT (called "Range") but I'll give him a general thumbs up for his work on advancing DIPS and note that he did a nice rundown on the strengths and weakness of various defensive stat systems under discussion here: http://www.hardballtimes.com/main/article/evaluating-the-evaluators/

201 Excellent once again!

But to answer a simple question like: Mchdcjkv sucks goat balls? Isn't RATE just fine?

201 yes, given a sample size of a couple hundred games, RATE offers a decent enough thumbnail assessment to assert that.

Though I'll wager that anyone administering the all-important Goat Balls Litmus Test of Defense to a particular player is well past the point of needing numbers to back up his/her argument.

203 Absolutely. The Goat Balls Test, once employed, sort of makes the conclusion moot.

Unfortunately, the folks polishing the glove of Mgsdfk won't acccept defensive metrics if they got to that point. So all you're really left with is the Goat Balls Test.