Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Tuesday, September 06, 2011

IATM: Is WAR the new RBI?

Give me OBP, give me OPS, give me IPO, give me WPA, give me K/BB; just don’t give me RBI!  If you’re going to give me RBI, Mr. McCarver, I’d rather you gave me nothing.

And then came WAR.

The concept was ratified by the sabremetric Godfather, Bill James, who’d created Win Shares according to a similar ideology in 2002.  It was a neoclassical economist’s wet dream, like baseball GDP: an elegant equation which accounted for all the sport’s diverse variables and yielded a single number roughly reducible to the oldest and most hallowed statistic of them all, the win.  Hallelujah.

Wins Above Replacement is a beautiful idea.  Euclidean grace in a quantum world.  A simple answer, not only for age-old baseball conundrums like “Mantle or DiMaggio?”, but also a formula for unprecedented comparisons like “Rickey Henderson v. Johnny Bench” and “Roy Halladay v. Alex Rodriguez“.

There’s only one problem.  It doesn’t work.

At least, not yet.  Not in the fantastically straight-forward way we try to use it.  The idea is so good, so clarifying – like democracy or the rational market – that we really, really want it to work, we’re willing to suspend our disbelief just a little while longer in the hope that it might.

...The cruel irony, the I-could’ve-had-Sean-Doolittle-and-all-I-got-was-stupid-Barry-Zito irony, is that the problem with WAR is the same as the problem with RBI.  It frequently measures context as much as performance.  Especially when used to evaluate single seasons, it doesn’t sufficiently account for the inevitable variations in opportunity and environment.

Thanks to Tango...

Repoz Posted: September 06, 2011 at 03:18 PM | 164 comment(s) Login to Bookmark
  Tags: history, projections, sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 2 of 2 pages  < 1 2
   101. Harold can be a fun sponge Posted: September 07, 2011 at 07:14 AM (#3918419)
"Or I could choose my own way to measure value points and decide that it's better."

And you can do the same for hits.


Also, thanks for quoting me out of context.
   102. whoisalhedges Posted: September 07, 2011 at 01:22 PM (#3918466)
I loves me some WAR, but of course it's not without its problems. Defense, yeah, the false sense of specificity... not to say that Win Shares is a better metric, but at least James told us flat out that a difference of +/- 3 WS in a season is insignificant, could just be statistical noise -- that's the kind of attitude we need to have when we look at WAR. Hell, that's the kind of attitude we need when we look at oRAR, probably significantly more wiggle room (and contextualizing, etc., etc.) with dRAR.

To suggest that someone with a 4.6 WAR is definitively superior to a player at 4.3 is asinine. To suggest that the 7.3 player is better than the 3.6, not so much. Even taking defensive regressions into account, even if it's "really" more like 6.2 vs. 4.7, there's a clear winner there. ;)

I love the concept of WS and WAR. I like the fact that there are people trying to define player contributions in terms of wins -- after all, we all want our team to win. The problems (and even the ones that haven't been mentioned in this thread: 10 runs per win, where's that come from? Methinks that can't be consistent across eras... also, defense at 3B and 1B was more prized a century ago, how's that accounted for -- it might be, but I dunno) don't mean that we (and by "we" I mean "not me, but someone better at maths) should abandon our search for the Holy Grail, it just means we haven't found it yet.

The admitted hyperbole in the article -- WAR is *not* RBI. It's not as deceptive. 17 runs on D for Carlos Lee? Ludicrous. But his season WAR does not misrepresent his value nearly as much as Joe Carter's 115 RBI in 1990. El Caballo *is* having a better season than Carter did in '90. Now, he ain't been worth no $18 million, of course.... I guess any stat is only as deceptive as you let it be.
   103. Matt Clement of Alexandria Posted: September 07, 2011 at 01:40 PM (#3918470)
"People keep saying this as if it's obvious, but I'm still unclear on what kind of actions I can take if I boil every player's contribution down to a single number."

Right. I see the former post, and wonder if this person has ever had to make any real decisions; and if so, does he do it by reducing everything to a single number?

OK, there are times at work where we do have to reduce a set of variables into a single value (total cost of ownership for a technology buy, for example); but then *we use a framework/variables specific to that decision*. We don't blindly use an all-purpose framework for any real decision.
I think it's useful here to take a look at the "player value" section in Baseball-Reference. This is Dustin Pedroia:

27 RBat, 4 Rbaser, -2 Rroe, 2 Rdp, 12 Rfield, 3 Rpos, 21 Rrep, 67 RAR, 6.5 WAR - Pedroia

And obviously you have his previous seasons, too. If you were making a decision about Dustin Pedroia, this is the information you'd want to have as to his value, its arrangement, and its measurement. You'd surely want other information, too, from his health to his personality to the aging curve of second basemen, to all the scouting reports that help you measure his value and projected value in ways the stats can't. But you'd also want to be looking at the value of the different aspects of his game in a form that allows for comparison between baserunning, fielding, hitting, and durability.

The WAR stats aren't just a single number. At B-Ref - and at Fangraphs - they present player value broken down into components, with explanations of how those components are calculated. It's really useful. It's not the end of a conversation, but it's almost always a good place to start.
   104. Lassus Posted: September 07, 2011 at 02:01 PM (#3918483)
The WAR stats aren't just a single number. At B-Ref - and at Fangraphs - they present player value broken down into components, with explanations of how those components are calculated. It's really useful. It's not the end of a conversation, but it's almost always a good place to start.

It is going to be a long, long time before anyone cites WAR to a large crowd - say, on television - and has this understood. And this is only slightly longer than the time that passes before someone actually attempts to make this understood, and this is really the main problem. Even bigger than the problem that gets under my skin about the defensive metric aspects of WAR.
   105. Matt Clement of Alexandria Posted: September 07, 2011 at 02:24 PM (#3918500)
Well, when you see a specific case of WAR being used improperly, that's a good time to complain. It doesn't appear to me that WAR is to be blamed for the inadequacy of modern, mainstream sports discourse.

The important thing to me is whether the stat is useful. Because of its presentation of component values in comparable forms, I find WAR to be the best first place to go when thinking about player value. So, I find it highly useful.

If what you want in a stat is a radical critique of mainstream sports discourse, or the tools for that radical critique, WAR probably falls short. If what you want in a stat is the readable and detailed presentation of useful information, WAR is one of the best we've got.
   106. Never Give an Inge (Dave) Posted: September 07, 2011 at 02:25 PM (#3918504)
Right. I see the former post, and wonder if this person has ever had to make any real decisions; and if so, does he do it by reducing everything to a single number?

OK, there are times at work where we do have to reduce a set of variables into a single value (total cost of ownership for a technology buy, for example); but then *we use a framework/variables specific to that decision*. We don't blindly use an all-purpose framework for any real decision.


You're arguing with a straw man. I never said you should always use an all-purpose framework. I just said that relying only on raw data and ignoring any interpretive metric that has uncertainty to it will not get you very far.

Sure, I have to make decisions, and usually those decisions are based on imperfect information. That doesn't mean I only rely on raw data to make them. A lot of the work I do relates to valuation of assets and companies, which involves taking a lot of uncertain data and variables and ultimately boiling them down into a single output--value or price. But at the same time, I try to understand and communicate the amount of uncertainty in my numbers, and I look at how sensitive my answer is to a number of underlying variables. And then I might look at ways to manage that risk, etc.
   107. Rickey! No. You move. Posted: September 07, 2011 at 02:28 PM (#3918506)
It seems to me that the problems people are mentioning here are problems with how (some people) use WAR.


I'll come back to this in just a minute.

What I don't agree with is when people (or perhaps just one person in this case) essentially say "I don't know why you'd want to do that anyway, I'll just make some adjustments in my head." Because with WAR, we can debate whether the defensive components are accurate, or whether the positional adjustments or replacement level are set to the right values. We can debate and discuss these things because by quantifying them we've made things specific, opened the process to scrutiny and (eventually) made the process verifiable (at least to a limited degree.) By discussing things this way our knowledge is enhanced, whereas ad-hoc personal adjustments can't be scrutinized in the same way.


And that is your fundamental error, I believe. You think that by "quantifying them" you've "made things specific" and thus, somehow, magically created an "open process to scrutiny" and "eventually made the process verifiable." All of this is false. All of this is hubris. Quantifying your mental adjustments doesn't make them more specific. Only numeric. But the real error is the notion that by opening up your "quantified" assumptions you're moving towards some sort of verified "truth." You're not. You're counting the number of angels on the head of a pin. And no amount of angel-counting will ever enhance "knowledge."

You're attempting to reduce complexity, because you want a counting number to describe loopy, unkempt reality. The error isn't in the specifics of your process, per se. The error is in attempting to shove the hairy beast of reality into your pristine little counter process in the first place.

Going back to your open, "It seems to me that the problems people are mentioning here are problems with how (some people) use WAR," this basically concedes the argument of the original article - that WAR is the new RBI. Because the problem with RBI wasn't with RBI, but with how (some people) use RBI." The problem with RBI is that a lot of people attempt to use that stat to reduce a complex interplay of multiple things - players on base in front of a hitter; the on-base player's speed and ability to score from first on doubles; the hitter's SLG ability; etc - into a single counting stat. The problem with WAR is that it attempts to reduce a complex interplay of multiple things - a player's skill at bat; his skill in the field; his place in the lineup; the composition of his teammates; the roster management tendencies of his manager - all of these enormous factors that create "wins" in the real world, to a single counting stat.

You're better off just sticking to slash stats (which are themselves overly precise, truth be told), noting era, league, park, position (where appropriate) and living with uncertainty on a number of questions. The drive for precision has over-ridden the basic truth of the game. That's no good.
   108. Matt Clement of Alexandria Posted: September 07, 2011 at 02:41 PM (#3918523)
The problem with WAR is that it attempts to reduce a complex interplay of multiple things - a player's skill at bat; his skill in the field; his place in the lineup; the composition of his teammates; the roster management tendencies of his manager - all of these enormous factors that create "wins" in the real world, to a single counting stat.
Perhaps. The good thing about WAR is that its presentation breaks up all of those components and displays them for you. You can say, "if we assume that 3Bs in the 70s were somewhat more valuable than they were in other decades, then Sal Bando has been underrated" and "if we set replacement level at around a .280 wOBA for a middle infielder, then Marco Scutaro is a reasonably valuable hitter" and "if Dustin Pedroia is an excellent defensive player, as these metrics suggest and as my observation agrees, then he's a real MVP candidate."

I agree that we need to recognize that we don't know baseball facts with the sort of precision that some uber-stats can appear to provide. However, I just don't understand where you think we should take baseball discussion. If you eyeball a line and make some mental adjustments, and I do the same thing, and we disagree - what's the next step? How do we settle our disagreement?

We could walk away and do something useful with our lives, or we could attempt to lay out in more detail how those mental adjustments worked. Only once we have some sort of shared language and grammar for discussion our mental adjustments can we have a real discussion. Stats like WAR and VORP and such are useful because they provide each one possible shared language for that discussion. We make a mistake when we assume that just one way of making adjustments is the right one, or that said adjustments are terribly precise, but without a shared language of some sort, we're not going to be able to have discussions about this.

You could say you don't want to have discussions about player value, and that's fine, but for people who do want to have those boring conversations, things like WAR are useful.
   109. BDC Posted: September 07, 2011 at 02:42 PM (#3918524)
I'm obviously of MCoA's opinion here, and I think it all comes down to what questions you're asking. If you want to know something about Dustin Pedroia, "6.5" tells you something extremely approximate. The components tell you more, and watching the guy every day while hoovering up everything that B-Ref and Fangraphs have to say about him tells you way more.

If you want to compare Pedroia to Bobby Doerr, or all Red Sox second basemen to all Yankee second basemen over the years, WAR is wonderful. Always to be elaborated with more specific data and with other perspectives, but it's the best thing going.
   110. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 02:44 PM (#3918526)
You're better off just sticking to slash stats (which are themselves overly precise, truth be told), noting era, league, park, position (where appropriate) and living with uncertainty on a number of questions. The drive for precision has over-ridden the basic truth of the game. That's no good.

So, you note that one player put up a 250/320/400 in 1968 Dodger Stadium playing a bad 2B w/ 12 SB/10 CS and bad baserunning, and another put up a 310/390/500 in 1997 Coors field playing a good 1B with 2SB/0CS and good baserunning, and another put up a 270/340/450 in 2010 Petco playing a great RF with 30 SB/5 CS and elite baserunning.

How do I go about comparing those w/o some quantification of the relative scale of adjustments needed for era, park, baserunning, defense and position? By just noting them, all I have is a series of +/- w/o any context as to magnitude.

Do the adjustment for me off the cuff. I'd like to see it.
   111. Rickey! No. You move. Posted: September 07, 2011 at 02:55 PM (#3918539)
Do the adjustment for me off the cuff. I'd like to see it.


You're not going to like this, but it's an honest response.

Why am I even trying to compare these two players at all?
   112. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 03:03 PM (#3918546)
You're not going to like this, but it's an honest response.

Why am I even trying to compare these two players at all?


HoF debate? Because we like to talk about baseball?

Or put all three into the same year, and call them FAs, and you want to know how much your team should bid on each.

It may be honest, but it's a fairly silly reponse. Baseball fans have been talking about "was Ruth better than Cobb", etc. forever.
   113. Matt Clement of Alexandria Posted: September 07, 2011 at 03:04 PM (#3918547)
If you don't care about discussions of player value, such that you have no interest in making such a comparison, then you're not going to need any shared language or grammar for discussing player value with other people. That's fine. No one has to be interested in that.

For people who do want to engage in these sorts of discussions, a shared language is highly useful, and WAR is one of the better options out there.
   114. Rickey! No. You move. Posted: September 07, 2011 at 03:07 PM (#3918549)
HoF debate? Because we like to talk about baseball?


HOF Debate - the Coors baby 1B who can't manage a SLG above 500 isn't a HOFer. A 720 OPS out of a 2B from '68 Dodger Stadium has a case to be made at least. Didn't bring much to the table defensively or as a base runner, so it's probably borderline barring any non-performance related HOF credentials (i.e. Jackie Robinson, etc.)

EDIT - A player active in Petco in 2010 is not eligible for the HOF.
   115. Ray (RDP) Posted: September 07, 2011 at 03:13 PM (#3918561)
I agree that we need to recognize that we don't know baseball facts with the sort of precision that some uber-stats can appear to provide. However, I just don't understand where you think we should take baseball discussion. If you eyeball a line and make some mental adjustments, and I do the same thing, and we disagree - what's the next step? How do we settle our disagreement?


Some disagreements can't be "settled." (Welcome to life.) You present your arguments and if they're both reasonable, that's that. Sometimes we can't know The Truth. But people using WAR pretend to. Or think they are.

Mind you I have no problem with the use of WAR in a discussion such as the above. If it's a reasonable argument, it's a reasonable argument. I just don't think it presents the final word on anything. (And, yes, now you'll tell me that nobody is presenting it as the final word. But they are.)
   116. Rickey! No. You move. Posted: September 07, 2011 at 03:14 PM (#3918563)
Or put all three into the same year, and call them FAs, and you want to know how much your team should bid on each.


What are the strengths and weaknesses of my team? What are my internal option for any given position? What are the secondary players available on the market?
   117. Never Give an Inge (Dave) Posted: September 07, 2011 at 03:29 PM (#3918571)
Quantifying your mental adjustments doesn't make them more specific. Only numeric.

The point of using these statistics is not to make the observations more specific, it's to make them more objective and consistent. Sometimes the consistency is unwarranted--for example, boiling down park factors to a single number may be useful for measuring retrospective player value versus a baseline, but it doesn't tell you how a specific player will do when moving to a new park.

But having some sense of the actual park factors helps us evaluate players, and avoids the problem where different people have wildly different views of, for example, how big a park adjustment is appropriate for Colorado. As MCoA said, it gives you a shared language and helps establish areas of agreement and disagreement in player evaluation. Maybe that's not something you're interested in; such is your prerogative.

But I wonder why you're using slash lines. Is a double really twice as good as a single, as SLG says? Is a walk really as good as a single, as OBP says? There's a certain false precision there too, at least if you're trying to use those stats in any way beyond their literal definitions (i.e. Albert Pujols reaches base safely in X% of his at-bats).
   118. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 03:30 PM (#3918572)
EDIT - A player active in Petco in 2010 is not eligible for the HOF.

OK, Mr. pedantic.

What are the strengths and weaknesses of my team? What are my internal option for any given position? What are the secondary players available on the market?

Your entire team is made up of replacement level players, as is the rest of the FA pool.

C'mon!
   119. Matt Clement of Alexandria Posted: September 07, 2011 at 03:33 PM (#3918575)
Some disagreements can't be "settled." (Welcome to life.) You present your arguments and if they're both reasonable, that's that. Sometimes we can't know The Truth. But people using WAR pretend to. Or think they are.
This reads way more into "settled" than I intended. I meant "settle" in an ongoing, having-a-discussion sense, not a determine-the-truth sense.

The question is, if we do as Sam suggests, and "make mental adjustments to slash lines", and then we have a disagreement about player value, what is the next step in the discussion? It has to be some way of communicating to each other the sorts of adjustments we made to the players' slash lines for park and league context, for baserunning and defense, for durability and so on. That communication requires some sort of shared language for talking about these adjustments. Something like WAR provides one set of vocabulary for the discussion - I made this sort of adjustment for playing time based on this sort of replacement level calculation, and I made this sort of adjustment for defense based on these defensive numbers and these observations, and I made this sort of adjustment for park based on these calculated park factors, and so on. The fact that we can communicate with each other the sorts of adjustments we made doesn't mean that either I or you has perfect knowledge of baseball - no one does. What it means is that we can communicate with each other in a reasonably comprehensible way.
   120. Never Give an Inge (Dave) Posted: September 07, 2011 at 03:36 PM (#3918578)
Some disagreements can't be "settled." (Welcome to life.) You present your arguments and if they're both reasonable, that's that.

If there was an award for irony, this post would win it.
   121. Matt Clement of Alexandria Posted: September 07, 2011 at 03:36 PM (#3918580)
Sam - are there comparisons across positions, across eras, between parks, or between players with different levels of durability that you are interested in?

If there are, then just lay one of those out for us and explain how you make the comparison in that case. If there aren't, then say that and explain that you don't care about these sorts of comparisons.
   122. Rickey! No. You move. Posted: September 07, 2011 at 03:37 PM (#3918581)
But I wonder why you're using slash lines.


Because each component of the slash line measures something real and useful.

Is a double really twice as good as a single, as SLG says?


Probably not in all cases. A single from Michael Bourn may actually be more useful to scoring a run than a double by Brian McCann. But it's close enough, and it tells me what the player did at least, if not how he did it.

Is a walk really as good as a single, as OBP says?


Absolutely not. But again, it's close enough. Precision isn't really that important in these things.

There's a certain false precision there too, at least if you're trying to use those stats in any way beyond their basic definitional uses.


I use slash stats to give me an idea of a player's offensive skill set. Any other use seems rather specious to me.
   123. Rickey! No. You move. Posted: September 07, 2011 at 03:42 PM (#3918586)
Your entire team is made up of replacement level players, as is the rest of the FA pool.

C'mon!


No, it's not. No team in the history of baseball has ever existed in such a state. My team, personally, has two super prospect/rookies in RF and 1B, an aging HOFer at 3B, two stables full of excellent pitching prospects, a low avg, moderate OBP, high SLG 2B, the best C in baseball, Martin Prado and Alex Gonzalez. They have few SS or LF prospects in the high minors, so in that case the 2010 Padre OFer who put up solid numbers in the death-zone of Petco looks like a good FA pursuit.

I'd probably not offer contracts to either the Coors 1B or the 1968 Dodger infielder, regardless of their respective HOF credentials. Because they're friggin' retired.
   124. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 03:50 PM (#3918593)
No, it's not. No team in the history of baseball has ever existed in such a state. My team, personally, has two super prospect/rookies in RF and 1B, an aging HOFer at 3B, two stables full of excellent pitching prospects, a low avg, moderate OBP, high SLG 2B, the best C in baseball, Martin Prado and Alex Gonzalez. They have few SS or LF prospects in the high minors, so in that case the 2010 Padre OFer who put up solid numbers in the death-zone of Petco looks like a good FA pursuit.

I'd probably not offer contracts to either the Coors 1B or the 1968 Dodger infielder, regardless of their respective HOF credentials. Because they're friggin' retired.


Obfuscation worthy of the worst Andy/Ray/David 500 post argument.

It's a F***ing hypothetical.

If you never wanted to compare two players across wildly different contexts, you're not much of a baseball fan I guess.
   125. Rickey! No. You move. Posted: September 07, 2011 at 04:00 PM (#3918605)
It's a F***ing hypothetical.


It's a *stupid* hypothetical that has no use value in the real world. Which is why you chose it. In order to defend a statistical flattening of complex reality, you must generate a hypothetical use case that does not exist in said reality. That's the entire f***ing point, son.

If you want me to compare two players across contexts, give me the players. Don't give me a hypothetical that is so absurd it could never exist in reality.

You put together a set of three theoretical "players" and then asked me to compare them for HOF cases. I point out that one of those players is, by your description, still active and thus ineligible for the HOF and you cry foul. The foul is on you, not me. If you want to compare HOF credentials, at least make the damned thought experiment logical.

Similarly, if you want me to discuss potential FA acquisitions for a given team, you have to provide some real context for team need and requirement. No team in the history of the game has ever had a blank slate where the players you made up have equal value. The reason I'm not playing your little theory game is because my entire point is that your theory is utterly detached from any actual reality.
   126. Ray (RDP) Posted: September 07, 2011 at 04:09 PM (#3918612)
Is there a place where all of the details of how bWAR is calculated are explained? I can't seem to find this, either on Sean's site or on b-r.
   127. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 04:14 PM (#3918618)

It's a *stupid* hypothetical that has no use value in the real world. Which is why you chose it. In order to defend a statistical flattening of complex reality, you must generate a hypothetical use case that does not exist in said reality. That's the entire f***ing point, son.


I chose it b/c it doesn't freaking matter, and I didn't want to spend 20 min on it.

The point is baseball fans want to compare players across disparate environments and skills sets on a consistent basis all the time. A framework like WAR (not necessarily WAR) is necessary to do that intelligently.

If you refuse to admit that those comparisons occur, there's no point in discussing it.

point out that one of those players is, by your description, still active and thus ineligible for the HOF and you cry foul.

Since when do we exclude active players from our HoF discussions? Hmmm, I must've seen a 1000 post Ichiro thread around here somewhere.
   128. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 04:15 PM (#3918620)
No team in the history of the game has ever had a blank slate where the players you made up have equal value. The reason I'm not playing your little theory game is because my entire point is that your theory is utterly detached from any actual reality.

Expansion teams enter the expansion draft with an effective blank slate with regard to MLB talent.
   129. Rickey! No. You move. Posted: September 07, 2011 at 04:16 PM (#3918622)
Since when do we exclude active players from our HoF discussions? Hmmm, I must've seen a 1000 post Ichiro thread around here somewhere.


None of which were based on a single slash line from a single season.
   130. Rickey! No. You move. Posted: September 07, 2011 at 04:17 PM (#3918624)
Expansion teams enter the expansion draft with an effective blank slate with regard to MLB talent.


Expansion teams have organizations in place. They know if they're good at scouting and developing pitching vs hitting. They're not blank slates.
   131. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 04:33 PM (#3918643)
None of which were based on a single slash line from a single season.

Pure sophistry. Again a simple hypothetical does not require that you have to provide 100% of the context. Take the lines to be the super-duper accurate ZiPs projection of what they'd do over their next contract.

The point is the need for a systematic method of comparison to reveal what we know, what we don't and what we might disagree on.
   132. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 04:34 PM (#3918644)
Expansion teams have organizations in place. They know if they're good at scouting and developing pitching vs hitting. They're not blank slates.

Based off one amateur draft. BS.

Your defense is getting painful to watch.
   133. Rickey! No. You move. Posted: September 07, 2011 at 05:06 PM (#3918673)
No, Snapper. You just don't understand that you're arguing idiocy to start with. You provided an idiotic, unreal "comparison." I pointed out where rational, reality based commentary would ask better questions before doing any sort of comp between your three "players." You then spun off into vitriol because you didn't like the fact that I pointed out that your original experiment was stupid on its face.
   134. Tom Nawrocki Posted: September 07, 2011 at 05:08 PM (#3918674)

If you never wanted to compare two players across wildly different contexts, you're not much of a baseball fan I guess.


I don't think the two sides here are really that far off. You're presenting WAR as a tool for comparing players in different contexts, mostly in the guise of a Hall of Fame vote; in other words, it's a fun thing for fans to play around with. Some people like to pretend that WAR is the cutting edge of baseball analysis, something that GMs and managers and announcers must pay heed of if they're to do their jobs properly, but it's not.

It's a fun thing for fans to play around with. It's the Favorite Toy.
   135. Tom Nawrocki Posted: September 07, 2011 at 05:13 PM (#3918679)
But I wonder why you're using slash lines. Is a double really twice as good as a single, as SLG says? Is a walk really as good as a single, as OBP says?


You're making Sam's case for him. Yes, we know a walk isn't really as good as a single; that's why we look at slugging percentage in addition to OBP. We're trying to aggregate information, not mash it up into a single ball. More information is always better than less.

Slash lines provide you with three distinct pieces of information about a player's hitting, and on top of that we're not stupid; we know that doesn't tell the whole story. WAR presents you with a single piece of information, with highly dubious (and ultimately unknowable) accuracy.
   136. Matt Clement of Alexandria Posted: September 07, 2011 at 05:20 PM (#3918684)
You're presenting WAR as a tool for comparing players in different contexts, mostly in the guise of a Hall of Fame vote; in other words, it's a fun thing for fans to play around with.
It's a tool for comparing players in different contexts, which is useful for HoF arguments, for assessing offseason moves and strategy, for any of the thousands of baseball arguments on the internet about who's better than who. It's a tool for talkin' about baseball, which we all like to do.

Sam is arguing that it is not a useful tool for talkin' about baseball. That's precisely his argument, and precisely where myself and snapper disagree with him. Based on your second post, I can't tell where you stand in that debate.
Some people like to pretend that WAR is the cutting edge of baseball analysis, something that GMs and managers and announcers must pay heed of if they're to do their jobs properly, but it's not.
If GMs and managers and announcers don't have a good handle on basic issues of player value - the value of different offensive events, park and league adjustments, positional adjustments, fielding, durability - then I think they're not doing their jobs as well as they should be. There's no need for them to use Fangraphs or CHONE WAR as the specific tool by which they talk about or analyze baseball value, but the sorts of analyses that comprise WAR are definitely things that people in baseball ought to be using.

EDIT: You said "cutting-edge" in that post, but no one here has said WAR is cutting-edge. None of the ideas which comprise WAR is less than 20 years old, I'd guess. It's just a good aggregator of data and analysis, presented intelligently and free of charge.
   137. Matt Clement of Alexandria Posted: September 07, 2011 at 05:23 PM (#3918688)
WAR presents you with a single piece of information, with highly dubious (and ultimately unknowable) accuracy.
No, WAR presents you with all of its many components laid out in a nice table ready to be copied into whatever spreadsheet you want. It's presented as many pieces of information, and you can manipulate them easily based on any new information you want to add to improve accuracy, account for uncertainty, or test hypotheses.
   138. PreBeaneAsFan Posted: September 07, 2011 at 05:28 PM (#3918695)
I'm going to post in this thread one more time before it gets to a level of nastiness that makes further discussion pointless.

First, I'm not going to bag on anyone for being interested/uninterested in making player comparisons. Some people find discussions comparing people who played different positions in different contexts interesting and other people don't. If you don't find such comparisons interesting you're not going to like WAR or any other uberstat. I don't think there's anything wrong with that. I don't really appreciate the derision towards people who are interested in such comparisons (sometimes we don't need a "real world" reason to do something and the fact that it is interesting and fun for us is enough), but I also don't agree with being nasty towards people who aren't interested in such comparisons.

However, I still feel like a number of people are missing an important point here. For example, I see one person saying:


But we don't really know when we have the *right* values, and what is *right* today might not have been *right* 25 or 30 years ago and may not be *right* 25 or 30 years from now.


another one saying

Some disagreements can't be "settled." (Welcome to life.) You present your arguments and if they're both reasonable, that's that. Sometimes we can't know The Truth.


I'll concede the point that we don't *know* when the adjustments we make in WAR are accurate. I'll go one step further and say that they are definitely wrong. Since when did being wrong about baseball become such a big crime? Since when did the impossibility of *knowing* The TRUTH become an excuse not to try? In all of science, economics, psychology, history and pretty much anything that isn't mathematics or logic this is the case, but that doesn't mean we throw our hands up and walk away. Our attempts to understand are informative. Our failures are informative. As MCoA stated, WAR gives us a common vocabulary and framework for having player value discussions. How much better does an average fielding 1B have to be with the bat than an average fielding 2B to have equal value? How does the value of an elite SP compare to an elite position player? Such comparisons may seem abstract, and actual team decisions will be based on many factors, but these kinds of relative valuations and discussions are important for understanding how to allocate your resources to construct a winning team. And after undergoing such a process of comparisons we walk away with more knowledge of how things work, even if we never attain a high level of certainty. That's how we advance our knowledge of the game. If you don't think that's an important thing to do, then by all means don't do it. If you have a better way to do it, then I'd love to hear it. But until then I'll find uses for stats like WAR.
   139. The District Attorney Posted: September 07, 2011 at 05:28 PM (#3918696)
The good thing about WAR is that its presentation breaks up all of those components and displays them for you. You can say, "if we assume that 3Bs in the 70s were somewhat more valuable than they were in other decades, then Sal Bando has been underrated" and "if we set replacement level at around a .280 wOBA for a middle infielder, then Marco Scutaro is a reasonably valuable hitter"
And when WAR is cited in a discussion, how often does everyone involved in the discussion understand that those are the underlying assumptions to be either defended or criticized?
   140. Matt Clement of Alexandria Posted: September 07, 2011 at 05:32 PM (#3918699)
And when WAR is cited in a discussion, how often does everyone involved in the discussion understand that those are the underlying assumptions to be either defended or criticized?
I don't know, but all discussion involves the use of tools and terminology which are based on reasoning which can be critiqued. Why is the fact that WAR is a tool for analysis like any other tool such a strike against it? WAR is in fact a pretty good tool based on pretty good reasoning, and it's presented in a way that allows for us to understand its workings reasonably well and make our own adjustments with little extra effort.
   141. Rickey! No. You move. Posted: September 07, 2011 at 05:33 PM (#3918700)
If GMs and managers and announcers don't have a good handle on basic issues of player value - the value of different offensive events, park and league adjustments, positional adjustments, fielding, durability - then I think they're not doing their jobs as well as they should be. There's no need for them to use Fangraphs or CHONE WAR as the specific tool by which they talk about or analyze baseball value, but the sorts of analyses that comprise WAR are definitely things that people in baseball ought to be using.


And barring outliers - hello Kansas City, how are you today? - they're doing this. Some of them do it better than others. Some of them do it in ways that piss off fans of a particular mindset more than others. But they're doing this. The problem isn't with doing that sort of critical thinking about players and teams. The problem is the growing trend of people who replace that sort of critical thinking with rote repetition of WAR tables, as if that's an argument in and of itself.
   142. The District Attorney Posted: September 07, 2011 at 05:37 PM (#3918706)
it's presented in a way that allows for us to understand its workings reasonably well and make our own reasoned adjustments with little extra effort.
When you use the passive-voice "it's presented", do you mean it's "presented" that way on BB-Ref, if you want to go through a lot of work to dig it out? Because it's not "presented" that way by the people actually using it in discussion.

That may not be "WAR's fault", but it is a hell of a problem with the discussion, no?
   143. Matt Clement of Alexandria Posted: September 07, 2011 at 05:40 PM (#3918711)
The problem is the growing trend of people who replace that sort of critical thinking with rote repetition of WAR tables, as if that's an argument in and of itself.
I don't read any other baseball blogs, so perhaps there's a crisis of WAR misuse out there that I am unaware of. I have seen very few BBTF threads where "rote repetition of tables" has been presented as an argument in and of itself. I use WAR and its components in arguments all the time, because of its utility as I think I've beaten to death in the posts about, and I think it's quite clear when I do so that I'm citing a tool based on certain kinds of reasoning, and if there are aspects of that reasoning which might be critiqued in a specific case, I consider those critiques and their possible upshot. It's very easy to use WAR in this sort of reasonably critical way, precisely because it is so intelligently arranged as being made up of commensurate analyses, and because it is so intelligently presented in multiple columns and rows to be analyzed.
   144. Matt Clement of Alexandria Posted: September 07, 2011 at 05:43 PM (#3918713)
When you use the passive-voice "it's presented", do you mean it's "presented" that way on BB-Ref, if you want to go through a lot of work to dig it out? Because it's not "presented" that way by the people actually using it.
WAR is presented on the Fangraphs and B-Ref websites with lots of component data, and it takes only the most nominal "work" to fire up a browser, type in a player's name to google, click on a link, and scroll down half a page. "A lot of work to dig it out"? Seriously?

If people have "dug out" this data and then presented it unintelligently, even though it was presented to them in an intelligent fashion, that's their fault. Again, perhaps out there on the blogs and the twitters there are people misusing WAR in egregious ways, with no one offering proper critiques, but I haven't seen this epidemic around these parts.
   145. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 05:47 PM (#3918714)
When you use the passive-voice "it's presented", do you mean it's "presented" that way on BB-Ref, if you want to go through a lot of work to dig it out?

Yes. 2nd table on the main player page. No digging required.
   146. SoSH U at work Posted: September 07, 2011 at 05:47 PM (#3918715)
Since when did being wrong about baseball become such a big crime?


I thought it was when fans like us started posting about the sport online.
   147. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 05:52 PM (#3918721)
Since when did being wrong about baseball become such a big crime?


The same time that umpires blowing a call became a threat to our way of life that must be dealt with by any means necessary.
   148. Rickey! No. You move. Posted: September 07, 2011 at 05:54 PM (#3918724)
The same time that umpires blowing a call became a threat to our way of life that must be dealt with by any means necessary.


So Ruth, or Cobb?
   149. The District Attorney Posted: September 07, 2011 at 05:55 PM (#3918728)
"A lot of work to dig it out"? Seriously?
I think it's fair. If I hit "Leaders" on Fangraphs, I get a list of guys sorted by WAR, with no breakdown of the components. Just like B-Ref popularized OPS+ by making it prominent in the stat display design, Fangraphs has done the same for WAR. Both sites have a #### ton of stats if you want to find them, but the presentation is such that certain stats are featured and others are not.

If people have "dug out" this data and then presented it unintelligently, even though it was presented to them in an intelligent fashion, that's their fault.
That's fine, and I figured it's what you would say, but:

1. It certainly is a point worth making, and emphasizing, that they are doing that.

2. It's not just certain people who cite the total rather than the components. It's EVERYBODY IN THE WORLD.

3. And honestly, it's really inevitable that that would be the case, that a stat designed to summarize a player's contribution in one number would be used by people as such. The discussion has become self-refuting: WAR is useful because it's convenient to sum everything up in one number, but you're not using it intelligently if you just use the one number.

4. Even if you were to tell me the components of Sal Bando's WAR, how would that tell me the thing that we apparently* need to know in order to have an intelligent discussion of Sal Bando's WAR, which is "if we assume that 3Bs in the 70s were somewhat more valuable than they were in other decades, then Sal Bando has been underrated"?

*I say "apparently" because I had no idea WAR was making this assumption until you said it just now.
   150. Ray (RDP) Posted: September 07, 2011 at 05:56 PM (#3918729)
I again ask where I can find a detailed breakdown of how WAR components are calculated. I don't see it on Sean's site or on b-r.

If the answer is that this information is not public, then that's the answer. Is that the answer? Is this stuff black box?
   151. AROM Posted: September 07, 2011 at 05:57 PM (#3918730)
The problem is the growing trend of people who replace that sort of critical thinking with rote repetition of WAR tables, as if that's an argument in and of itself.


Could you link to an example of this?
   152. Tom Nawrocki Posted: September 07, 2011 at 05:57 PM (#3918732)
WAR presents you with all of its many components laid out in a nice table ready to be copied into whatever spreadsheet you want.


Maybe it's just semantics, but I would disagree with this. Baseball-reference.com or Fangraphs presents you with all of WAR's many components laid out in a nice table ready to be copied into whatever spreadsheet you want. WAR, by definition, mashes all those components together in an attempt to come up with a single statistic that represents a player's value.

Maybe you intend to mean "the WAR process" or some such, but I don't think the WAR process is necessary for people to want to come up with all those component stats. Fans would likely want to figure how many runs a player's GIDP were costing his team whether there was a WAR or not.
   153. AROM Posted: September 07, 2011 at 06:04 PM (#3918744)
If the answer is that this information is not public, then that's the answer. Is that the answer? Is this stuff black box?


Somewhere in between. Totalzone was originally written on MVN Statistically speaking, which no longer exists. There are some articles on the system in the THT archives, just click on my name and I think it was in 2007.

Any explanations of batting runs, baserunning, etc. might be found in posts on my old angels blog, or a thread here, or on Tango's site. Not exactly well organized, I know. It's not black box, just hard to find.

The best I can do for you on my site is this:
http://www.baseballprojection.com/war/statdef.htm

I'm willing to spend some time and explain in detail exactly what goes into all of the calculations. But you'll have to find a book deal willing to give me a decent advance first.
   154. Matt Clement of Alexandria Posted: September 07, 2011 at 06:10 PM (#3918748)
The discussion has become self-refuting: WAR is useful because it's convenient to sum everything up in one number, but you're not using it intelligently if you just use the one number.
WAR is useful because it's useful, not just convenient, to have component aspects of player value presented in commensurate form.
Maybe you intend to mean "the WAR process" or some such, but I don't think the WAR process is necessary for people to want to come up with all those component stats. Fans would likely want to figure how many runs a player's GIDP were costing his team whether there was a WAR or not.
Perhaps, but before Fangraphs and B-Ref started publishing WAR, there was no good place to go to get most of the information presented on those pages now. I don't think anyone had a single baserunning component number listed (maybe BP?), and I know no one had a single ROE or GDP component rating listed. Further, while we had stats like VORP, we did not have any places that broke VORPish numbers into their three components (batting value above average, batting value above replacement, positional value).People may have wanted to figure out how many runs a player's GIDP was costing his team - in fact, people did want it, I wanted it - but before WAR, we didn't have a good place to get it.
I think it's fair. If I hit "Leaders" on Fangraphs, I get a list of guys sorted by WAR, with no breakdown of the components.
Hit "value". A second click still doesn't amount to "a lot of work to dig it out."
   155. Rickey! No. You move. Posted: September 07, 2011 at 06:28 PM (#3918766)
Could you link to an example of this?


www.bravesjournal.com
www.capitolavenueclub.com

I'm relatively certain that Mac's and Peter's blogs are not outliers.
   156. Never Give an Inge (Dave) Posted: September 07, 2011 at 06:29 PM (#3918767)
You're making Sam's case for him. Yes, we know a walk isn't really as good as a single; that's why we look at slugging percentage in addition to OBP. We're trying to aggregate information, not mash it up into a single ball. More information is always better than less.

"Aggregating" versus "mashing up" is sort of a meaningless distinction for purposes of this discussion as nobody here is arguing for "mashing up" (as in losing or ignoring information in the process of analyzing). What people are saying is that WAR or similarly designed stats and their components can provide you with additional valuable analysis in addition to what the raw stats tell you. Not that you should use WAR and nothing else.
   157. zenbitz Posted: September 07, 2011 at 09:15 PM (#3918927)
Getting Old sucks. But it beats the alternative.
   158. snapper (history's 42nd greatest monster) Posted: September 07, 2011 at 09:20 PM (#3918934)
So Ruth, or Cobb?

Ruth of course. Did you have to ask?
   159. Rickey! No. You move. Posted: September 07, 2011 at 10:02 PM (#3918953)
Ruth of course. Did you have to ask?


No, but Cobb could have, if he'd wanted to.
   160. Never Give an Inge (Dave) Posted: September 07, 2011 at 10:40 PM (#3918972)
www.bravesjournal.com
www.capitolavenueclub.com

I'm relatively certain that Mac's and Peter's blogs are not outliers.


What am I missing? I looked at the front pages of both those sites. One was just a bunch of game recaps and the other had 5-6 posts, including two Transaction Analysis posts, none of which mentioned WAR.

I'm not doubting that these authors may have misused WAR at some point in the past, but every stat is misused by someone. I don't see why that should dissuade those who advocate using them properly.
   161. Eric J can SABER all he wants to Posted: September 07, 2011 at 10:49 PM (#3918976)
There are some articles on the system in the THT archives, just click on my name and I think it was in 2007.

The Brian Downing vs. Jim Rice article from 2009 also gives a reasonable summation of the method.
   162. TDF, situational idiot Posted: September 07, 2011 at 11:24 PM (#3919003)
I think the problem is that on top of the yearly rollercoaster

I was just referring to the people who say the yearly roller coaster means there is a problem.
I'm guessing this is directed at my Jay Bruce example.

That's not my point. My point is that since 3 years of data is needed, where on the "roller coaster" is Bruce's real ability? Any one year is an inherently unreliable sample, but we treat that one year sample as significant in calculating WAR. We use as accurate what would be the same as using just 2 months' worth of hitting data to calculate the offensive portion of WAR.

By both fangraphs and bbref, Bruce's defense is a full 2 wins worse than last year, and a win worse than 2 years ago. AROM says "But is a 4.4 player better than 3.5? Not absolutely, not to the extent we can be sure Barry Bonds was better than Royce Clayton, but very probably." My point is that in the case of Jay Bruce, we can't (not don't, but can't) know if he's a 1.6 WAR player (as defined by bbref), 2.6 (if the average of the last 3 years is close to his true defensive value), or 3.6 (if last year was his true defensive value).
   163. TDF, situational idiot Posted: September 07, 2011 at 11:24 PM (#3919004)
EDIT: the dreaded double post.
   164. Rickey! No. You move. Posted: September 08, 2011 at 01:05 AM (#3919084)
I'm not doubting that these authors may have misused WAR at some point in the past, but every stat is misused by someone. I don't see why that should dissuade those who advocate using them properly.


It's more the commentors than the owners, though Peter does love him some WAR.
Page 2 of 2 pages  < 1 2

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Support BBTF

donate

Thanks to
phredbird
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

Sox TherapyJust Keep It Rolling
(26 - 11:42pm, Jun 25)
Last: Dillon Gee Escape Plan

NewsblogIt's all OMNICHATTER and lollipops from here on out, for June 25, 2017
(82 - 11:36pm, Jun 25)
Last: Meatwad

NewsblogDetroit Tigers release struggling closer Francisco Rodriguez
(25 - 11:36pm, Jun 25)
Last: Dillon Gee Escape Plan

Sox TherapyMinor Moves
(6 - 11:31pm, Jun 25)
Last: Dillon Gee Escape Plan

NewsblogOT - March 2017 NBA thread
(7209 - 11:28pm, Jun 25)
Last: Athletic Supporter wants to move your money around

NewsblogYankees designate Chris Carter for assignment, call up Tyler Austin from Triple-A
(17 - 11:15pm, Jun 25)
Last: SoSHially Unacceptable

NewsblogOTP 19 June 2017: Bipartisan baseball: Dems best GOP, give trophy to wounded Scalise
(1061 - 10:50pm, Jun 25)
Last: Ray (RDP)

NewsblogColumn | Freddie Freeman Shouldn't Play 3B To Accommodate Adams
(41 - 10:32pm, Jun 25)
Last: cardsfanboy

NewsblogKenley Jansen vs Craig Kimbrel as top reliever | MLB.com
(11 - 10:21pm, Jun 25)
Last: cardsfanboy

NewsblogDodgers' Julio Urias to have shoulder surgery; he's out for the season
(20 - 10:05pm, Jun 25)
Last: Covfefe

NewsblogRecounting the many ways baseball has changed in 20 years
(74 - 9:32pm, Jun 25)
Last: PreservedFish

NewsblogCENTRAL PARK - JULY 1, 2017 10 AM - BBTF ANNUAL SOFTBALL EXTRAVAGANZA
(271 - 8:55pm, Jun 25)
Last: Howie Menckel

NewsblogTim Tebow moves up a level, proving Mets only care about one thing
(16 - 8:55pm, Jun 25)
Last: Adam Starblind

NewsblogJorge Mas To Offer $1.1 Billion For Miami Marlins
(2 - 7:15pm, Jun 25)
Last: Bote Man

NewsblogCreeping Coachism
(2 - 6:52pm, Jun 25)
Last: Walt Davis

Page rendered in 0.5035 seconds
49 querie(s) executed