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

Sunday, May 18, 2014

Bob Ryan: Do baseball fans care about new breed of stats?

Or as Paul O’Neill said…“Shifts might get you outs, but they won’t save you any runs”. Or something.

My question is, does the average person care? Is the average fan still content with batting average, runs batted in, and earned run average being the Holy Trinity of baseball stats, even though the modern Smart Guys have discredited all three? Oh, and — how could I forget? — wins. Speak not to the modern baseball analysts about a pitcher’s wins, those being the most circumstantial of pitching developments, at least in their eyes.

I’m guessing that most fans are oblivious to all the new statistical stuff. They just want to watch and enjoy a game. They will continue to evaluate players and teams by giving everyone the Eye Test, just as their father, grandfather, and great-grandfather did. If this means they are then wallowing in some kind of statistical ignorance, then so be it. I think the average fan really didn’t understand the recent fuss over whether Miguel Cabrera was worthy of an MVP. He won the Triple Crown in 2012, didn’t he? Isn’t the Triple Crown supposed to be baseball’s crowning offensive achievement? Hadn’t we been waiting since 1967 to see another one? Of course, Miguel Cabrera was worthy of being MVP. Next question.

And didn’t he put up even better numbers last year? He should have been the MVP a second time, and he was.

But wait. The New Breed Stat Guys were apoplectic in both 2012 and 2013. They stopped short of declaring Cabrera a fraud, but they did say he was unworthy because, after all, he was out-WARPed by Mike Trout in both seasons, 10.9-7.3 in 2012 and 8.9-7.5 last season. End of story, as Tony Soprano might say.

For those of you old-timers who don’t know, WARP (or WAR), stands for “Wins Above Replacement Player,” and it is the be-all and end-all number for the New Breed Stat Guys. As defined in Baseball Prospectus (an indispensable preseason guide even diamond Luddites can enjoy), WARP attempts to quantify “the difference between the ‘replacement level’ (derived from looking at the quality of players added to a team’s roster after the start of the season) and the league average.” You are allowed to remain skeptical about its validity in my view, but the New Breed Stat Guys swear by it.

What ultimately matters is whether you can still appreciate a given baseball game. I wonder if the New Breed Stat Guys ever actually enjoy a game, because they are so obsessed with what the manager is or isn’t doing, based on the data in front of them. They’re often upset before the game even starts, because the lineup isn’t sufficiently stat-based. And God forbid the skipper who doesn’t properly handle what they have termed “high leverage” situations. Sometimes lost in all this is an appreciation of the aesthetics, whether it’s a great play in the hole by a shortstop or a snappy inning-ending 5-4-3 double play or a base runner cleverly taking an extra base. Or even a game-winning hit in the ninth inning if it happens to be delivered by someone other than the guy they thought should have been up at the plate. Sometimes the New Breed Stat Guys aren’t so good about accepting the vagaries of a very complex game.

Repoz Posted: May 18, 2014 at 04:35 PM | 79 comment(s) Login to Bookmark
  Tags: 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.

   1. Publius Publicola Posted: May 18, 2014 at 04:50 PM (#4709202)
What ultimately matters is whether you can still appreciate a given baseball game. I wonder if the New Breed Stat Guys ever actually enjoy a game, because they are so obsessed with what the manager is or isn’t doing, based on the data in front of them.


Mr. Average Baseball Fan is just as, if not more so, obsessed with second-guessing the manager. That's been happening since time immemorial.

Sometimes the New Breed Stat Guys aren’t so good about accepting the vagaries of a very complex game.


Wow. Talk about lack of self-awareness.
   2. BDC Posted: May 18, 2014 at 04:58 PM (#4709210)
You are allowed to remain skeptical about {WAR's} validity in my view, but the New Breed Stat Guys swear by it


As Primates know, it's usually the opposite. Hardly an analytical thread goes by here without somebody interrogating some aspect of the validity of WAR. It's mainly deriders of any new stats at all who assume that WAR is now some kind of article of the New Faith.

As to "does the average person care" – well, the average person probably doesn't care about baseball. There is also a slice of fans who just want to wear the team colors and cheer home runs and strikeouts and that's great, enjoy life! The Stat Police are not patrolling the aisles enforcing their allegiance to FIPs.

But if you frame it as "does the average person who's inclined to embark on arguments over baseball and cite statistics care," I actually think yes. It's no different than when I was a little kid and I would say that Ernie Banks had hit a lot of home runs but somebody else would tell me Billy Williams had a better batting average. Much later I learned that Ron Santo had a better OBP and was probably at that point better than either of them :-D When I was nine, I took in pretty quickly that Bob Gibson was better than Denny McLain because he had an incredible ERA, even though McLain had won 30 games. Education begins somewhere, and some people take to it.

So, most people aren't going to get a doctorate in this stuff. I know a very little bit about advanced stats, about 1% of what most of y'all know, but I hang out here to learn more. And so it goes: on the scale of how much you care at a given moment, you're always at the point of caring more.

   3. valuearbitrageur Posted: May 18, 2014 at 04:58 PM (#4709211)
What I enjoy most about watching a baseball game is that usually I can't hear or see Bob Ryan while doing so, and thus don't have to concern myself with the opinions of a blowhard.
   4. Jose Is The Most Absurd Thing on the Site Posted: May 18, 2014 at 05:25 PM (#4709221)
As Primates know, it's usually the opposite. Hardly an analytical thread goes by here without somebody interrogating some aspect of the validity of WAR. It's mainly deriders of any new stats at all who assume that WAR is now some kind of article of the New Faith.


I don't know about that. For every person around here who questions WAR it seems like 90% of arguments are reduced to "WAR says player X is superior." I think there probably IS an over reliance on WAR.
   5. Howie Menckel Posted: May 18, 2014 at 05:42 PM (#4709225)
agreed re WAR, but the point about Trout is just that he's a much better baserunner and fielder than Cabrera, so even a dunce should be able to grasp that it cuts into Cabrera's advantage. Saying, "not by enough" is one thing. ignoring those aspects entirely is abject stupidity.
   6. Bring Me the Head of Alfredo Griffin (Vlad) Posted: May 18, 2014 at 05:50 PM (#4709230)
Bob Ryan is still alive? When did that happen?
   7. Gonfalon Bubble Posted: May 18, 2014 at 05:53 PM (#4709233)
I thought "The Sports Reporters" was one of those traveling exhibitions of posed and embalmed corpses.
   8. Dock Ellis on Acid Posted: May 18, 2014 at 06:09 PM (#4709244)
Sometimes I forget I live in a bubble. I hang out on here, I read sites like Fangraphs and my baseball fan friends and I only care about RBI when it concerns our fantasy teams. WAR is also the point of reference when I water-cooler talk baseball with the guys in the office. So when I hear something like, "But he's only batting .265!" I probably make the same face I would when someone insists that the world is flat. I don't think I really associate with any baseball fans who don't care about post-Moneyball metrics, so I honestly have no idea just how mainstream this is.

   9. Ray (RDP) Posted: May 18, 2014 at 06:26 PM (#4709247)
The title should have been "Does Bob Ryan Care About the New Breed of Stats? No. No, He Doesn't. RBIs and Pitcher Wins Forever!"
   10. DavidFoss Posted: May 18, 2014 at 06:34 PM (#4709249)
There seems to be much less fixation on the "batting race" these days.

It used to be that all summer long, you'd know who led the league in batting average. With the exception of 2012's Triple Crown race, it seems to be something you notice on the final weekend. Whereas people generally still know who leads in HR's all summer long.

Could also be a side effect of the internet replacing newspapers. The leaders box in the corner of the page would catch my eye all summer, now seeing the leaders requires an extra click.
   11. Captain Supporter Posted: May 18, 2014 at 07:01 PM (#4709256)
As Primates know, it's usually the opposite. Hardly an analytical thread goes by here without somebody interrogating some aspect of the validity of WAR. It's mainly deriders of any new stats at all who assume that WAR is now some kind of article of the New Faith.


Funniest thing I have ever read on this site. This site basically consists of people who evaluate people based on WAR. See any of the interminable Hall of Merit posts if you even want to try to disagree. Or see most of the posts in the last two years re: Cabrera vs. Trout.

WAR is a useful but imperfect stat. And it appears that it used that way by actual major league baseball teams, even 'progressive'ones. It is not anything resembling the ultimate arbiter of all baseball disputes, as it so often used by people with a superficial knowledge of the new metrics. WAR does give you an argument as to why Mike Trout could have been the MVP. But it in no way gives you anything resembling a reason to argue (as many on this site vociferously did) that anyone who disagrees with that thesis is an ignorant slug unaware of the sublime insights provided by "post moneyball metrics".
   12. the Hugh Jorgan returns Posted: May 18, 2014 at 07:31 PM (#4709265)
I’m guessing that most fans are oblivious to all the new statistical stuff.


I love stuff like this. That arguments about whether stats enhance the experience for the fan are silly. With any endeavor, there is always a group of people who want to explore the hows, whens and whys of things. I always liken it to economics.
According to Bob Ryan everyone in North Korea has a job...the economy must be great! People like ourselves just want to look at some numbers and see if that's actually true. Except of course in North Korea we'd be executed for asking those questions, so there is that.
   13. bobm Posted: May 18, 2014 at 07:33 PM (#4709266)
Bob Ryan writes like the only stat he cares about is blood alcohol content.
   14. Joey B. Posted: May 18, 2014 at 08:08 PM (#4709279)
Funniest thing I have ever read on this site. This site basically consists of people who evaluate people based on WAR.

And the damnedest thing about it is that there's no such thing as a universal WAR formula. The last time I checked, there were still two of them!
   15. valuearbitrageur Posted: May 18, 2014 at 08:13 PM (#4709280)
WAR is a useful but imperfect stat. And it appears that it used that way by actual major league baseball teams, even 'progressive'ones. It is not anything resembling the ultimate arbiter of all baseball disputes, as it so often used by people with a superficial knowledge of the new metrics. WAR does give you an argument as to why Mike Trout could have been the MVP. But it in no way gives you anything resembling a reason to argue (as many on this site vociferously did) that anyone who disagrees with that thesis is an ignorant slug unaware of the sublime insights provided by "post moneyball metrics".


MVP is the most valuable player, which WAR, for all it's minor imperfections, is perfectly suited for. And anyone on the Cabrera side of the MVP discussion was a lazy slug, the differences in value were far to great to dismiss because of problems in WAR. You literally had to believe that defense didn't matter, base running didn't matter, positional value didn't matter, hitting into double plays didn't matter, all that mattered was batting average, home runs, and your teammates OBP. And that the world remained flat.

Except of course in North Korea we'd be executed for asking those questions, so there is that.


Some of you would just be rumoured to be executed, and I would trot you out every few years just to prove my benevolence and my detractors wrong.
   16. Hank G. Posted: May 18, 2014 at 08:21 PM (#4709282)
And the damnedest thing about it is that there's no such thing as a universal WAR formula. The last time I checked, there were still two of them!


There are at least three. WARP is WAR under another acronym.

Tom Tango claims that everyone who makes a judgement about which of two players is better is using WAR, albeit a personal version that might not be consistent or codified.
   17. valuearbitrageur Posted: May 18, 2014 at 09:27 PM (#4709296)
Tom Tango claims that everyone who makes a judgement about which of two players is better is using WAR, albeit a personal version that might not be consistent or codified.


that's a key point, GMs have been using WAR for over 100 years, they just didnt use a math framework
   18. Baldrick Posted: May 18, 2014 at 11:39 PM (#4709322)
I’m guessing that most fans are oblivious to all the new statistical stuff. They just want to watch and enjoy a game. They will continue to evaluate players and teams by giving everyone the Eye Test, just as their father, grandfather, and great-grandfather did. If this means they are then wallowing in some kind of statistical ignorance, then so be it. I think the average fan really didn’t understand the recent fuss over whether Miguel Cabrera was worthy of an MVP. He won the Triple Crown in 2012, didn’t he? Isn’t the Triple Crown supposed to be baseball’s crowning offensive achievement? Hadn’t we been waiting since 1967 to see another one? Of course, Miguel Cabrera was worthy of being MVP. Next question.

Ah yes, the famous Eye Test. Where you simply use your Eyes to evaluate the player('s three specific stats that Bob Ryan cares about).

Is there any conceivable world where baseball players had to be evaluated purely based on the Eye Test (where NO statistics of any kind were kept) that wouldn't have Mike Trout as the two-time reigning MVP? I mean, Miguel Cabrera is great but Mike Trout is the Platonic Ideal of a baseball player by the Eye Test.
   19. cardsfanboy Posted: May 19, 2014 at 12:14 AM (#4709331)
I don't know about that. For every person around here who questions WAR it seems like 90% of arguments are reduced to "WAR says player X is superior." I think there probably IS an over reliance on WAR.


I think you are overstating peoples adherence to War. Sure there are some who stick to it, including accepting that 3.2 war over 160 game season is correctly better than a 2.8 war over 160 game season. But most people around here know
1. War is not useful past the decimal point
2. Which components are legitimately arguable, while accepting(for the most part) the etched in stone data.

The only thing that surprises me in any war debate around here, is that there are actually people who use dWar in talking about players from two different positions and quality of their defense.

agreed re WAR, but the point about Trout is just that he's a much better baserunner and fielder than Cabrera, so even a dunce should be able to grasp that it cuts into Cabrera's advantage. Saying, "not by enough" is one thing. ignoring those aspects entirely is abject stupidity.


Even though the war difference between Trout and Cabrera was greater in 2012(3.6) than 2013(1.4)....I think there is a much more legitimate argument for Cabrera in 2012 than there is in 2013.

Funniest thing I have ever read on this site. This site basically consists of people who evaluate people based on WAR. See any of the interminable Hall of Merit posts if you even want to try to disagree. Or see most of the posts in the last two years re: Cabrera vs. Trout.


Then you have never ever read this site. War is a tool, that is useful for compiling all the known data that a player participates in, into one number. That is pretty much it, it gives a nice eyeball test to see how much value a player provided, but very few people on here use it as the only part of the debate, just as a point to "numberize" their debate.


WAR is a useful but imperfect stat. And it appears that it used that way by actual major league baseball teams, even 'progressive'ones.


There is literally nothing in war that a baseball team should be using. They need to work with the components, any team that ever says "we took this guy because he posted a 4.0 war last season, and we think that 2 wins above our previous guy at that position is enough to get us over the hump" is an organization that is going to be hanging around with the Royals for the next decade. (or whoever happens to be the sad sack team) War is a value stat. A great centerfielder on a team with two crappy corner outfielders is going to get a larger defensive bonus to his individual War stat than if he was on a team with two good corner outfielders. Using an individual's War and trying to project that player on another team is just pure folly.


MVP is the most valuable player, which WAR, for all it's minor imperfections, is perfectly suited for. And anyone on the Cabrera side of the MVP discussion was a lazy slug, the differences in value were far to great to dismiss because of problems in WAR. You literally had to believe that defense didn't matter, base running didn't matter, positional value didn't matter, hitting into double plays didn't matter, all that mattered was batting average, home runs, and your teammates OBP. And that the world remained flat.


I agree with the first part, and am on the Trout side of the discussion, but absolutely 100% disagree with the second part(part about Cabrera's defenders being lazy...which is true often times, but not always...in fact many of Trout defenders were equally guilty of being lazy in not questioning the War components) . Trout's difference in 2012 over Cabrera was aided by suspect defensive factors, suspect park factors, suspect base running factors, and completely misses the narrative of the team side of the equation that Cabrera did by accepting a defensive hit to put more bats into the team. And of course the nearly 20 games difference in playing time, that Trout made up a lot in plate appearances, but that was helped a little bit by lineup spot. Etc.

And the damnedest thing about it is that there's no such thing as a universal WAR formula. The last time I checked, there were still two of them!


I was under the impression that for the most part, both systems used the same formula, it's just that the components and weighing of them that are different.
   20. cardsfanboy Posted: May 19, 2014 at 12:19 AM (#4709332)
Is there any conceivable world where baseball players had to be evaluated purely based on the Eye Test (where NO statistics of any kind were kept) that wouldn't have Mike Trout as the two-time reigning MVP? I mean, Miguel Cabrera is great but Mike Trout is the Platonic Ideal of a baseball player by the Eye Test.


Yep, this is what bothers me most about the Trout/Cabrera debate... Trout is the guy that the old schoolers claim that the new schoolers wouldn't be able to appreciate, yet the roles are now reversed...but only on this argument.
   21. cardsfanboy Posted: May 19, 2014 at 12:39 AM (#4709336)
The other funny thing about the entire Cabrera/Trout debate is that the new school people, the stat obsessed guys that Ryan is attacking, are the ones who have been touting Cabrera for a lot longer than the old school guys, who for some reason can only acknowledge the quality of the player if he happens to be on a contender.

I guarantee you that you will find an(more) article(s) written in 2004/2005 propping Cabrera up by a Calcaterra or Neyer type long before you are going to see one by a Ryan or Murray type.
   22. cmd600 Posted: May 19, 2014 at 01:26 AM (#4709343)
.(JavaScript must be enabled to view this email address) Though I guess technically, the logic is consistent.
   23. KingKaufman Posted: May 19, 2014 at 02:16 AM (#4709349)
The thing that drives me crazy about pieces like this is that they're so incredibly journalistically lazy.

Aside from people who make a living out of disseminating and analyzing said data, who else pays attention? Just asking.


Yeah. JUST asking. Why don't you use your skills and years of training and experience and ####### ANSWER? Maybe go, I don't know, here's a crazy idea, TALK TO BASEBALL FANS and ask them if they pay attention. You don't even have to get your fat ass out of the chair. You could use the MASSIVELY POWERFUL WORLDWIDE NETWORK AT YOUR FINGERTIPS and do an unscientific survey of the thousands upon thousands of people who can read your words. You have 64,000 Twitter followers and there are a few people who read the Boston Globe. Or you could see about getting one of the hugely powerful media organizations you work for to commission a scientific poll.

My question is, does the average person care? Is the average fan still content with batting average, runs batted in, [etc etc]


Gee, how can this be known? Average people are so hard to find! It's a total unknown, what "the average person" thinks about or cares about. I mean, sure, you could look into how popular books, magazines, websites, etc. that talk about advanced baseball stats are. But how would you know how many of those people are average?

I’m guessing that most fans are oblivious to all the new statistical stuff. They just want to watch and enjoy a game.


Well by all means, guess. That's what the great journalists do. Woodward and Bernstein guessed, and that's how we know today about the crucial role played in the Watergate break-in by Soupy Sales.

I wonder if the New Breed Stat Guys ever actually enjoy a game, because they are so obsessed with what the manager is or isn’t doing, based on the data in front of them.


This really is a puzzle, Bobby Boo. I mean, how can one know this? It's not like these "New Breed Stat Guys" spend any time online, or would be willing to respond to an email or a tweet or a Facebook message from a famous Boston Globe columnist and on-air ESPN personality. It's not like there are "New Breed Stat Guys" in any press box, both working media members and front-office people. I mean, "New Breed Stat Guys" are as rare as Sumatran Tigers -- or average people.
   24. Baldrick Posted: May 19, 2014 at 02:22 AM (#4709351)
I wonder if the New Breed Stat Guys ever actually enjoy a game, because they are so obsessed with what the manager is or isn’t doing, based on the data in front of them.

Us New Breed Stat Guys sure do love to micromanage the managerial decisions. That's why we hate Earl Weaver so much.
   25. cardsfanboy Posted: May 19, 2014 at 02:35 AM (#4709354)
Thanks post 22...thanks for the try...but use the a tag and not the at tag. Here is the link from post 22.

Edit: I apologize if that sounded "bad", it wasn't meant to. As I appreciated the article.

From the article linked.

Regarding your NL MVP candidates, how about those two guys in Florida? Yes, the Marlins are not in playoff contention, but it's hard to ignore Hanley Ramirez and Miguel Cabrera, especially considering they're first and second, respectively, in the NL in VORP, and rank in the top three in Runs Created. It looks like you went through all the playoff-contending teams, and chose a "good" player from each. Let me ask you: If Cabrera were on a playoff-contender this season, would there be any doubt who the MVP was?
-- Carolyn, Boca Raton, Fla.

Actually, you're right. That's exactly what I did, and how I came up with Prince Fielder as my NL MVP leader. His "good'' year is actually more than good, and the Brewers are right in the thick of the playoff race. While I understand your sentiments, I am more interested in "wins created'' than runs created. And the day I consider VORP is the day I get out of the business. The idea of the MVP is to honor the player who has had the biggest positive impact on the pennant races. I have been a big champion for Ramirez, but I would not consider him a true candidate to win the MVP award.
   26. Gonfalon Bubble Posted: May 19, 2014 at 03:03 AM (#4709355)
Jon Heyman from the above link: "And the day I consider VORP is the day I get out of the business."

Open request to Jon Heyman: please consider VORP. Today.
   27. vivaelpujols Posted: May 19, 2014 at 03:24 AM (#4709356)
Some baseball fans care about new stats, the majority don't. Next question. You can stop pretending to be a moron now, Bob Ryan.

WAR is also a more accurate reflection of value than batting average or wins, but you might not care about that so whatever. And as far as I know no one thinks WAR is perfect.
   28. Cooper Nielson Posted: May 19, 2014 at 04:34 AM (#4709358)
MVP is the most valuable player, which WAR, for all it's minor imperfections, is perfectly suited for. And anyone on the Cabrera side of the MVP discussion was a lazy slug, the differences in value were far to great to dismiss because of problems in WAR. You literally had to believe that defense didn't matter, base running didn't matter, positional value didn't matter, hitting into double plays didn't matter, all that mattered was batting average, home runs, and your teammates OBP. And that the world remained flat.

I'm gonna echo cardsfanboy here... It was a pretty convincing difference in WAR in 2012*, but the Anaheim park factors over the past few years have seemed very suspect, the default positional adjustment still seems to need tweaking, and I'm not at all convinced that WAR measures defense accurately (by dWAR, Trout went from Devon White to Manny Ramirez and apparently back to Devon White in successive years -- seems unlikely). I realize that some of those work in Trout's favor and some do not; this is just an overall observation on the reliability of WAR.

I think WAR's a great concept and is generally accurate, but I'm not sure that differences of 2-3 WAR are necessarily definitive, and I'm quite certain that differences of 1 WAR are not.

* I'm a huge Cabrera fan, but I would've voted for Trout in 2012. However, 2013 seemed like a toss-up with at least 4 guys having legit MVP cases.
   29. Baldrick Posted: May 19, 2014 at 05:03 AM (#4709359)
2-3 WAR is a HUGE difference. That's the difference between an mediocre starter and an All-Star. If all you mean by 'not necessarily definitive' is that you can imagine a circumstance where a convergence of irregularities could conceivably close the gap, well sure. But I am confident to at least a 95% interval that a 2-3 WAR gap indicate a genuine and significant difference in value.

Similarly, I am quite confident that a difference of 1 WAR is meaningful. I obviously don't treat it as definitive, and am happy to dig deeper if there are arguments presented. WAR is a beginning to the conversation not the end of it. But if the only information I've got is that there was a 1 WAR gap between two guys, I am perfectly happy assuming that there is in fact a significant value gap. Sure it could be smaller. It could also be larger. WAR gives us an estimate, which isn't conclusive, but it's a lot better than a guess.
   30. Cooper Nielson Posted: May 19, 2014 at 07:40 AM (#4709367)
2-3 WAR is a HUGE difference. That's the difference between an mediocre starter and an All-Star. If all you mean by 'not necessarily definitive' is that you can imagine a circumstance where a convergence of irregularities could conceivably close the gap, well sure.

Yes, that's pretty much what I mean. If someone has 9.0 WAR and someone else has 7.0 WAR, that's surely a good indication that the first player is "more valuable," but I don't think it's an invitation to be adamant that "OF COURSE PLAYER 1 WAS BETTER AND YOU MUST BE AN IDIOT IF YOU DON'T BELIEVE THAT."

There will most likely be occasions -- especially when you're comparing very different positions (i.e., pitchers to designated hitters to defensive specialists) in very different contexts -- where the second player arguably provided as much value.

Even if ALL the currently hazy factors that go into WAR (park factors, defense, positional adjustments) were somehow guaranteed to be correct, the measure is still just an "estimate" of value. It's an idealized form of value that doesn't take into account what actually happened in terms of real-life runs or wins.

A guy who hits 10 solo home runs in the 9th inning of blowouts gets the same credit as a guy who hits 10 lead-assuming grand slams. (But the second guy's home runs contribute more to winning games.)

A guy who goes 0 for 10 with runners on 3rd base and 6 for 10 with the bases empty (and gets stranded at first every time) gets the same credit as a guy who goes 6 for 10 with runners on 3rd base and 0 for 10 with the bases empty. (But the second guy's team scores more runs in real life.)

A DH who goes 8 for 8 in a 21-0 win then goes 0 for 12 over his next three games gets the same credit as a DH who goes 2 for 5 in four straight games. But the first guy contributed zero value over the last three games, and as great as he was in the first game, it only resulted in one win. The second guy likely contributed more "value."

(Or am I wrong here? Apologies if I am misunderstanding how WAR works.)

Eventually, it should all even out, but over 120-162 games, this might not necessarily happen. I could easily see a swing of 0.5+ WAR either way based on the difference between "estimated value" (what ideally should have happened) and "real value" (what actually did happen). If you've got one WAR-lucky guy and one WAR-unlucky guy, and you add in the structural issues, it's certainly possible that a 2.1 WAR difference is not definitive.

WAR gives us an estimate, which isn't conclusive, but it's a lot better than a guess.

I wholeheartedly agree.
   31. Joey B. Posted: May 19, 2014 at 10:06 AM (#4709415)
I was under the impression that for the most part, both systems used the same formula, it's just that the components and weighing of them that are different.

They're obviously using different formulas, and those different formulas sometimes sometimes lead to very different results.

Just to give one example: according to FanGraphs, Stephen Strasburg has been one of the best pitchers in the N.L. so far, with a WAR of 1.5. Yet accoring to B-R, he has been nothing all that special, with a WAR of just 0.4

A difference of 1.1 strikes me as being pretty large this early in the season. They can't possibly both be right at the same time here. So which one of them is right?
   32. JL Posted: May 19, 2014 at 10:23 AM (#4709422)
agreed re WAR, but the point about Trout is just that he's a much better baserunner and fielder than Cabrera, so even a dunce should be able to grasp that it cuts into Cabrera's advantage. Saying, "not by enough" is one thing. ignoring those aspects entirely is abject stupidity.


What I have found fascinating about the sports writers attack on WAR is that it seems to under cut what they had previously fought for during the home run explosion.

When Bonds and the like were breaking the records, there was much bemoaning how baseball had become boring, it was like a slow pitch softball game and players were not learning the fundamentals. In many respects, Cabrera has continued that (though his defense at third base was always exciting in a "who knows what will happen" sort of way). Along comes Trout, who so clearly does a lot of those little things right. The advanced metrics take that into account and say that Trout, because he is good in a ton of areas, is actually the better player. And many of those writers who complained about station to station baseball now go on the attack - against an advance stat that seems to support what they had been saying. I just don't get it.
   33. Jeltzandini Posted: May 19, 2014 at 10:34 AM (#4709425)
WAR as a concept is what pretty much everybody has always done to evaluate players. Smith is a better hitter than Jones, does Jones make up the difference by fielding better at a harder position, and running the bases better. The calculations that go into WAR-as-stat are imperfect but will presumably improve.

This is somehow an enormous controversy.
   34. if nature called, ladodger34 would listen Posted: May 19, 2014 at 11:04 AM (#4709441)
What ultimately matters is whether you can still appreciate a given baseball game. I wonder if the New Breed Stat Guys ever actually enjoy a game, because they are so obsessed with what the manager is or isn’t doing, based on the data in front of them.


Mr. Ryan needs to come to site like this and hang out for a bit. I would say that anyone who posts or lurks here on a regular basis probably enjoys the game far more than most people in the country (too bad I didn't have a metric for that to drive Mr. Ryan crazy).

But for goodness sakes, even before I stumbled upon Neyer and followed a link in one of his columns over here, I still questioned the managers all the damn time. It's part of the game. Bob Ryan (this dumb column aside) is a walking basketball encyclopedia. I would bet that Mr. Ryan has questioned a basketball coach's decisions a gazillion times. If he says otherwise, he is a liar. And I know he loves basketball.
   35. Heinie Mantush (Krusty) Posted: May 19, 2014 at 11:44 AM (#4709477)
I think sabermetrics, beyond a few major things like OBP, WHIP, OPS, WAR, and the problems with pitcher wins, haven't really hit into the mainstream. I don't really find this that problematic, either. Most of the stats we use today will be considered relics in 5-10 years (the way blind adherence to FIP was foolhardy in 2006 or defense didn't matter to BPro in 2001.) I'd rather that stuff not become CW.

Today, WAR is the stat du jour and it's a good place to start a conversation because it's better than a guess. However, seconding CFB,I really really think that blind adherence to WAR is a) both more common and b) runs deeper in a more visceral way than it should in the sabermetric community at large (i.e. including the 17 year olds who post on /r/baseball.)

I think the WAR framework is fantastic. I think that it will eventually breakdown into context-dependent WAR and otherwise, etc. and with improvements in measuring park/defense/external factors, we'll get better at using it but seriously.. there's a *ton* of gray yet. I think the Fangraphs pitcher WAR treatment (i.e. RA9 vs FIP-WAR) is sort of the future, and I'd love to see an REW (or something) WAR for position players.
   36. villageidiom Posted: May 19, 2014 at 12:21 PM (#4709507)
I endorse #23. All hail the King.

- - - - -

Actually, I also endorse BDC's comment in #2 about Gibson and McLain. That's basically the best place to start if you want to explain the point of new stats to someone otherwise resistant: Who was better in 1968, Gibson or McLain? (And if they say McLain because wins, ask them about the 6/5/1968 game, in which McLain was removed from the game with his team trailing 4-2 and got the win.)

- - - - -

If anyone is convinced that the traditional stats are all you need, then offer to buy them lunch and they can explain their rationale over a burger.

And then take 'em to McDonald's. I mean, why have any other burger but the one that's been the world's best-selling burger for many decades? Why complicate the meal with some newfangled place that probably takes more time and costs more? It can't possibly be worth it.

   37. SoSHially Unacceptable Posted: May 19, 2014 at 12:37 PM (#4709523)
Who was better in 1968, Gibson or McLain? (And if they say McLain because wins, ask them about the 6/5/1968 game, in which McLain was removed from the game with his team trailing 4-2 and got the win.)


While I agree with most of what you're saying, I can't say that I ever remember anyone speaking about or writing in support of the idea that McLain's 1968 was better than Gibson's. Maybe it happened frequently in 1968, which was a little early for me, but since I've been following the sport Gibby's year is the one far more people point to in awe than the old scumbag's.

   38. villageidiom Posted: May 19, 2014 at 12:48 PM (#4709530)
While I agree with most of what you're saying, I can't say that I ever remember anyone speaking about or writing in support of the idea that McLain's 1968 was better than Gibson's.
If someone is making the argument that pitcher wins are the best statistic to measure pitcher performance, they're kinda forced into arguing McLain's 1968 was better than Gibson's. Otherwise, I don't know who would even attempt to argue it.
   39. Bring Me the Head of Alfredo Griffin (Vlad) Posted: May 19, 2014 at 12:54 PM (#4709533)
Mr. Ryan needs to come to site like this and hang out for a bit.


Please don't encourage him.
   40. SoSHially Unacceptable Posted: May 19, 2014 at 12:55 PM (#4709535)
I guess I see your point, though I think it would be terribly hard to find the hypothetical foil to provide that answer. Even among the strong believers in the Power of the W, I don't recall encountering anyone who takes it to that extreme or is completely oblivious to the run prevention side of the equation.

   41. if nature called, ladodger34 would listen Posted: May 19, 2014 at 12:56 PM (#4709536)
Please don't encourage him.


Why not? Just based on his basketball stuff, Bob actually does actually have the ability to understand this stuff better than he lets on.
   42. Shooty Is Disappointed With His Midstream Urine Posted: May 19, 2014 at 12:56 PM (#4709538)
I think the growing popularity of SB Nation is really bringing the new stats into the mainstream since most of those team blogs are sabr friendly. Of course you don't need stats to enjoy the game. I went to a game with my mom recently and she had a good time in her way and I did in mine. It was all good. As for Bob Ryan...change is scary. I don't think I've cared what Bob Ryan had to say on anything in at least 15 years if ever.
   43. Bring Me the Head of Alfredo Griffin (Vlad) Posted: May 19, 2014 at 01:13 PM (#4709550)
Why not?


The last thing I need in my life is some self-important blowhard telling me that I'm enjoying my baseball the wrong way.
   44. Lance Reddick! Lance him! Posted: May 19, 2014 at 01:18 PM (#4709554)
A guy who hits 10 solo home runs in the 9th inning of blowouts gets the same credit as a guy who hits 10 lead-assuming grand slams. (But the second guy's home runs contribute more to winning games.)

You're under no obligation to accept Rbat. Feel free to plug in Skoog/Ruane-approved value-added batting runs or whatever you'd like in its place.

Almost all WAR criticism boils down to "I don't agree with Component X." Then change it. Don't think a fielder bounces from +20 to -10 to +20 in consecutive years? Call him +10 every year. Or +12, +6 and +12. Whatever the hell you want. For every component on the list.
   45. if nature called, ladodger34 would listen Posted: May 19, 2014 at 01:39 PM (#4709565)
The last thing I need in my life is some self-important blowhard telling me that I'm enjoying my baseball the wrong way.


My point was, Bob should hang out here to see that we love baseball just as much (and probably more) than he does. If he was a jackass and said otherwise, so be it.
   46. Cooper Nielson Posted: May 19, 2014 at 01:44 PM (#4709570)
Almost all WAR criticism boils down to "I don't agree with Component X." Then change it. Don't think a fielder bounces from +20 to -10 to +20 in consecutive years? Call him +10 every year. Or +12, +6 and +12. Whatever the hell you want. For every component on the list.

How would that improve WAR? You seem to have mistaken me for an expert.

The guys who invented WAR know a lot more about this stuff than I do, and I trust them to "fix" (tweak/improve) it more than I would trust myself. It gets better every year. I still don't think it's perfect. What's controversial about that?
   47. PreservedFish Posted: May 19, 2014 at 01:54 PM (#4709577)
I'm a little surprised that some of you still have the capacity to get annoyed by an article like this. Didn't you get tired of making these same arguments a decade ago?
   48. if nature called, ladodger34 would listen Posted: May 19, 2014 at 02:01 PM (#4709584)
I'm a little surprised that some of you still have the capacity to get annoyed by an article like this. Didn't you get tired of making these same arguments a decade ago?


I really think the thing is that an article like this was sort of "okay" a decade ago. Now, it's just ridiculous. In fact, I think these are articles are more annoying now than they were back when I first stumbled upon this site.
   49. PreservedFish Posted: May 19, 2014 at 02:12 PM (#4709601)
I really think the thing is that an article like this was sort of "okay" a decade ago. Now, it's just ridiculous.


Yes. Our arguments a decade ago were good and passionate and they accomplished things. But now the Bob Ryans are just pathetic - the battle is over, they lost - bringing the same type of logic and standards to bear against their lazy columns is a waste of time, it seems to me.
   50. if nature called, ladodger34 would listen Posted: May 19, 2014 at 02:26 PM (#4709615)
Yes. Our arguments a decade ago were good and passionate and they accomplished things. But now the Bob Ryans are just pathetic - the battle is over, they lost - bringing the same type of logic and standards to bear against their lazy columns is a waste of time, it seems to me.


I get it, but Bob is bringing up the same lazy assed arguments from a decade ago. I think it is alright to continue to give these folks a little grief. Not get worked up over it, but I can still be a little taken aback when these folks want to claim that I don't enjoy a baseball game. The only reason I don't enjoy a baseball game is because my emotions get the best of me but that is only because I am overly invested in it.
   51. zenbitz Posted: May 19, 2014 at 02:56 PM (#4709638)
The real question: Do Stat fans care about the new breed of baseballer!
   52. SoCalDemon Posted: May 19, 2014 at 03:23 PM (#4709666)
I'm not at all convinced that WAR measures defense accurately (by dWAR, Trout went from Devon White to Manny Ramirez and apparently back to Devon White in successive years -- seems unlikely)


But this happens with offensive stats all the time, without people having these kind of qualms (I am not saying that the defensive stats are as good as the offensive stats, but I believe there is a lot of signal in the noise).

Melky Cabrera went from a 157 to 87 to 127 OPS+, seems unlikely.

Vernon Wells....seems unlikely. Morgan Ensberg went from a 131 to a 90 to a 144 OPS+, seems unlikely.

Adrian Beltre went from an 88 to a 163 to a 93 OPS+, seems unlikely.

And on and on and on....whether through luck or injury or true changes in ability, batters have wild swings in performance all of the time. I would not assume there is any difference in the field.
   53. Jose Is The Most Absurd Thing on the Site Posted: May 19, 2014 at 03:34 PM (#4709678)
And on and on and on....whether through luck or injury or true changes in ability, batters have wild swings in performance all of the time. I would not assume there is any difference in the field.


I think you are right but intuitively it feels wrong. We have over a century of history to easily grasp that offense tends to ebb and flow. Guys have breakout years/bad years all the time. Defense has been traditionally perceived as more constant so it seems wrong somehow that it would fluctuate.

I think it DOES fluctuate (though not as much as offense usually) but it just "feels" wrong.
   54. SoSHially Unacceptable Posted: May 19, 2014 at 03:41 PM (#4709688)
And on and on and on....whether through luck or injury or true changes in ability, batters have wild swings in performance all of the time. I would not assume there is any difference in the field.


I don't believe defense ebbs and flows nearly as much as offense, because there are simply more factors involved in producing offensive results.

With offensive performers, you have a) the pitchers trying to get you out, b) the umpires making calls on balls and strikes, and c) the defense actively trying to stop base hits. All of these factors, plus the fact that offensive players don't really have any say on where they hit the ball, just how hard, and you've got a recipe for fluctuation.

Defense is basically you and the ball, and the ball really doesn't care if you catch it. True talent should manifest itself far more directly in this circumstance.

Now, the flip side of that is that we're much better at measuring offensive results than we are defense. But when we reach a point where defensive stats are as trustworthy as offensive ones, I believe we'll see less variation on the defensive side of the ball (given truly comparable sample sizes).
   55. ellsbury my heart at wounded knee Posted: May 19, 2014 at 04:19 PM (#4709712)
Now, the flip side of that is that we're much better at measuring offensive results than we are defense. But when we reach a point where defensive stats are as trustworthy as offensive ones, I believe we'll see less variation on the defensive side of the ball (given truly comparable sample sizes).


I think I disagree. I think there's even more variation on the defensive side, if only because there are generally very few difficult chances from which to make judgments about player ability. 93% of of Jacoby Ellsbury's fielding opportunities last year were routine, according to Inside Edge. He had 24 opportunities to make non-routine plays. That's a really small number, and I think estimates from that are going to be inherently more unstable. Now you can look at routine chances, but when MLB players make 95% of those and there's only a few % points difference between ok and great, it just doesn't leave much variability from which to draw meaningful contrasts. Even if a player's true defensive talent is more stable, there are just fewer opportunities to observe it in a game, regardless of whether the metric used is trustworthy or not. I think that generally smaller sample is going to limit the fineness of distinction between players you can make with offensive stats. Not to mention the difference in number and quality of opportunities between a SS and a LF or something.

EDIT: Unless you're talking about MLBAM-type efficiency-of-route, process-oriented stats, in which case I might agree with you. I'm not sure I've formed an opinion other than "Wow!" on that stuff yet.
   56. Publius Publicola Posted: May 19, 2014 at 04:25 PM (#4709715)
I think defense is MORE prone to fluctuation than offense. Often, a guy nursing an injury that precludes him playing the field will still be used as a DH or pinchhitter. An outfielder with a hammy pull can still swing the bat OK but will lose a ton of range in the field.
   57. if nature called, ladodger34 would listen Posted: May 19, 2014 at 04:29 PM (#4709718)
I know my chances at getting a response are about 0%, but I did ask Mr. Ryan on Twitter if he ever questions an NBA coach's decision process during a game...
   58. SoSHially Unacceptable Posted: May 19, 2014 at 05:33 PM (#4709755)
I think I disagree. I think there's even more variation on the defensive side, if only because there are generally very few difficult chances from which to make judgments about player ability. 93% of of Jacoby Ellsbury's fielding opportunities last year were routine, according to Inside Edge. He had 24 opportunities to make non-routine plays. That's a really small number, and I think estimates from that are going to be inherently more unstable. Now you can look at routine chances, but when MLB players make 95% of those and there's only a few % points difference between ok and great, it just doesn't leave much variability from which to draw meaningful contrasts. Even if a player's true defensive talent is more stable, there are just fewer opportunities to observe it in a game, regardless of whether the metric used is trustworthy or not. I think that generally smaller sample is going to limit the fineness of distinction between players you can make with offensive stats. Not to mention the difference in number and quality of opportunities between a SS and a LF or something.


That's why I wrote, "given truly comparable sample sizes."

However, the fact that basically all major leaguers make all the plays, then we're really just measuring at the edges anyway. And I firmly believe, given the rather narrow band of skills that go into defense (essentially, positioning, jump and pure range), and the nature of these skills in terms of being able to repeat them, I believe that defensive skill will be less prone to fluctuation when the tests used to measure them are better calibrated.

   59. Moeball Posted: May 19, 2014 at 05:42 PM (#4709758)
Us New Breed Stat Guys sure do love to micromanage the managerial decisions. That's why we hate Earl Weaver so much.


Earl would be spinning in his grave right now if he knew that the Orioles were last in the major leagues right now in walks by their batters. They're the only team that hasn't reached 100 walks as a team yet.

Even my Padres with their woeful, last in the majors team .280 OBA (which is still up a dozen points from where it was a week ago - see what playing a few games at Coors Field can do for you?)have more walks than the O's.
   60. Johnny Sycophant-Laden Fora Posted: May 19, 2014 at 07:49 PM (#4709810)
Is there any conceivable world where baseball players had to be evaluated purely based on the Eye Test (where NO statistics of any kind were kept) that wouldn't have Mike Trout as the two-time reigning MVP?


This was an argument I made a long long time ago, if you didn't know anyone's AVG-HR-RBI and you simply watched Keith Hernandez and Steve Garvey play, NO ONE would have said Garvey was the better player, but until Bill James came along the vast majority of reporters and fans would have said Garvey was better.
   61. cardsfanboy Posted: May 19, 2014 at 07:54 PM (#4709814)
However, the fact that basically all major leaguers make all the plays, then we're really just measuring at the edges anyway. And I firmly believe, given the rather narrow band of skills that go into defense (essentially, positioning, jump and pure range), and the nature of these skills in terms of being able to repeat them, I believe that defensive skill will be less prone to fluctuation when the tests used to measure them are better calibrated.


Not just the edges, but the quality of the surrounding players on the team. Especially when it comes to centerfielders. There is an argument that there is really a cap of runs a players defense can save over an average player of his position, when you see players putting up higher numbers than the theoretical cap, it's not because of otherwordly defense, it's because of weak teammates that they are able to compensate for.(or selective play....say a third baseman with fantastic range like Rolen in his prime, is constantly taking balls that the average shortstop fields, this is going to make Rolen look greater than a thirdbaseman who allows the shortstop to field the ball, and it's going to hurt Rolen's shortstop, because now he has fewer plays made, and it makes mistakes have a stronger impact on his numbers, even though at the team level, the results were exactly the same.) In some respects, defensive stats can be similar to rbi, in that they can be team dependent.
   62. Rob_Wood Posted: May 19, 2014 at 07:57 PM (#4709817)

My analog is the fantasy-football owners I hang out with. When I first encountered them I thought it was a shame that they could not enjoy the NFL the way I did (I had been following the NFL for 20+ years). Well, when I hang out with these guys during football season, they know a hell of a lot more about what is happening in the NFL than I do (or ever did). They know every head coach's name and tendencies, every quarterback, every running back, every wide receiver, every kicker, every team's strengths and weaknesses, etc. If I were to try to suggest that these guys do not know football or do not enjoy football, they would laugh me out of the room (and they'd be right to do so).

While the analogy is not perfect, the analytical baseball community is similar. I am very confident that I and many others on this site are in the top 1% of people when it comes to watching baseball in terms of quantity (I will put the number of baseball games I watch in a week up against Bob Ryan any day) and enjoyment. Sure my enjoyment may not always be the same type of enjoyment as the non-analytical fan, but that is to be expected. At times during a game, and in post-game cogitation, I may focus on different things than other fans might, but generally speaking I imagine the vast majority of times we enjoy the same things: a dramatic hit, a majestic home run, a well-turned double-play, a well-pitched game, a good defensive play, a ten-pitch at bat that ends with a walk, a key strikeout, etc.

Good grief, these types of articles bother me, have always bothered me, and will always bother me.
   63. cardsfanboy Posted: May 19, 2014 at 08:24 PM (#4709833)

This was an argument I made a long long time ago, if you didn't know anyone's AVG-HR-RBI and you simply watched Keith Hernandez and Steve Garvey play, NO ONE would have said Garvey was the better player, but until Bill James came along the vast majority of reporters and fans would have said Garvey was better.


Going by their HOF results, I think that is about a true statement. Keith's best finish in the hof vote was 10.3 while Garvey's was 41.6...and this was in the mid 90's when people were making these votes.

The sad part is
Keith Hernadez .296/.384/.436/.821 128 ops+, 1071 rbi, 1124 runs, 162 hr....117 rfield, 8553 pa
Steve Garvey .294/.329/.446/.775 117 ops+, 1308 rbi, 1143 runs, 272 hr. 0 rField, 9466 pa.
(Garvey averaged 19 hr and 91 rbi per 162 games vs Keith 13 hr and 83 per 162..) So what it really boils down to, is roughly 8 rbi per year more, and 6 homeruns is why the writers think that Garvey was better, even though Keith's defense was otherworldly, better average and better obp.)

   64. Cooper Nielson Posted: May 19, 2014 at 11:35 PM (#4709916)
Not just the edges, but the quality of the surrounding players on the team. Especially when it comes to centerfielders. There is an argument that there is really a cap of runs a players defense can save over an average player of his position, when you see players putting up higher numbers than the theoretical cap, it's not because of otherwordly defense, it's because of weak teammates that they are able to compensate for.(or selective play....say a third baseman with fantastic range like Rolen in his prime, is constantly taking balls that the average shortstop fields, this is going to make Rolen look greater than a thirdbaseman who allows the shortstop to field the ball, and it's going to hurt Rolen's shortstop, because now he has fewer plays made, and it makes mistakes have a stronger impact on his numbers, even though at the team level, the results were exactly the same.) In some respects, defensive stats can be similar to rbi, in that they can be team dependent.

This is related to another thing I've been thinking about. Let's say you're the 2012-13 Angels and you have two outstanding centerfielders in Mike Trout and Peter Bourjos, with Bourjos considered to be the better defensive player by most evaluators. (I wish this wasn't Trout we were talking about, because then everyone thinks it's about Trout vs. Cabrera, but it's a good real-life example and I can't think of a better one at the moment.)

If Bourjos plays CF and Trout plays LF, that's theoretically going to result in better team defense, i.e., fewer runs surrendered by the Angels. But if Trout plays CF and Bourjos plays LF, that's better for Trout's WAR, because of the positional adjustment (even if he doesn't play CF quite as well as he plays LF). So Trout becomes a "more valuable" player simply because his manager makes a suboptimal decision. But his being "more valuable" leads directly to more Angels losses, I guess because he's stealing WAR from Bourjos.

If everything is perfectly calibrated (it probably isn't/can't be), then the increase in Trout's WAR should be more than offset by a decrease in Bourjos's WAR -- the combined WAR of Trout + Bourjos should drop a bit because the better defensive player is getting fewer opportunities to make value-added plays. It would be interesting to see if this is actually the case, but I suppose it would be hard to get a large enough sample size to run the experiment.
   65. PreservedFish Posted: May 20, 2014 at 12:17 AM (#4709937)
But if Trout plays CF and Bourjos plays LF, that's better for Trout's WAR, because of the positional adjustment

I believe that the thought behind the defensive adjustments is that Trout's WAR would remain the same no matter what position he played, exactly because he doesn't play CF as well as he plays LF.

That relies on an assumption that we know to be ridiculous, that all players, despite vast differences in skill sets, will experience equivalent changes in defensive production when switched from one position to any other ... but it's how the system is designed, and I can't think of a better way.
   66. Cooper Nielson Posted: May 20, 2014 at 12:40 AM (#4709944)
I believe that the thought behind the defensive adjustments is that Trout's WAR would remain the same no matter what position he played, exactly because he doesn't play CF as well as he plays LF.

But doesn't his offense also get a boost because now he hits better "for a center fielder" than he did "for a left fielder"? Also, his dWAR could improve because he's given more opportunities to make value-added plays, even if he's an "A-" centerfielder vs. an "A" leftfielder. (Moving him from CF to DH would surely decrease his WAR; I assume moving him from CF to 1B would too, in that his defense at 1B couldn't possibly be so exceptional that it made up for the higher offense expected at the position.)

For argument's sake, let's say that instead of Trout and Bourjos, the Angels had two identical Mike Trouts, one in CF and one in LF, with equal defensive ability and putting up identical offensive lines. Wouldn't the CF Trout have a higher WAR?

I am asking for enlightenment -- I actually don't know how it works.
   67. ellsbury my heart at wounded knee Posted: May 20, 2014 at 01:24 AM (#4709952)
I believe that defensive skill will be less prone to fluctuation when the tests used to measure them are better calibrated.


I guess I'm just curious what you think those tests will be. I think we've kind of reached the limits of precision with UZR-type metrics, and I don't see how they're going to get much better calibrated. I do agree that when/if we start getting those MLBAM-type goodies, defensive metrics are gonna go bananas.
   68. CrosbyBird Posted: May 20, 2014 at 01:35 AM (#4709956)
For argument's sake, let's say that instead of Trout and Bourjos, the Angels had two identical Mike Trouts, one in CF and one in LF, with equal defensive ability and putting up identical offensive lines. Wouldn't the CF Trout have a higher WAR?

I think it could well be the other way around. Because defense is compared to average at the position, a superstar LF might well be able to be much higher above the average defensively than he loses by being compared to the average LF rather than CF (which would be a positional adjustment in WAR).
   69. PreservedFish Posted: May 20, 2014 at 02:14 AM (#4709965)
For argument's sake, let's say that instead of Trout and Bourjos, the Angels had two identical Mike Trouts, one in CF and one in LF, with equal defensive ability and putting up identical offensive lines. Wouldn't the CF Trout have a higher WAR?

I'm not entirely sure myself. But I believe that the positional adjustments are supposed to account for this move and they would have the same WAR. That is to say, the LF Mike Trout would get dinged by X runs for playing the lesser position, and we also assume that he would be X runs better at that defending that position.

Would it actually work this way in real life? I don't know.
   70. Cooper Nielson Posted: May 20, 2014 at 04:33 AM (#4709969)
I think it could well be the other way around. Because defense is compared to average at the position, a superstar LF might well be able to be much higher above the average defensively than he loses by being compared to the average LF rather than CF (which would be a positional adjustment in WAR).

But then it gets to the issue that someone raised above about the "theoretical maximum" defense an outfielder can provide.

Let's say that instead of Mike Trout, we have young Andruw Jones or Willie Mays, in Fenway Park. Put that guy in left field, with a competent center fielder beside him, and he catches just about every flyball between the foul-line wall, the Green Monster, the infield dirt, and the center fielder, but that's still, what, 3 catches a game? How many non-routine chances is he going to get? There's a fairly low ceiling on "how much better" he can be than the average defensive left fielder.

Put him in centerfield, however, and he's got much more room to roam, plus the potential of an occasional over-the-fence grab, to show how much better than "average" he is.

Remember, we're not talking about an average outfielder being asked to play center. We're talking about a very capable centerfielder who just happens to play with an even better one. (Or in the twin Trouts case, an equal one.) It seems to me there are more non-routine opportunities in centerfield, and thus it would be possible for an excellent defensive outfielder to get higher above average at the position, even if the average is significantly higher. Or, in other words, the difference between a peak Andruw Jones (godlike CF) and a peak Torii Hunter (Gold Glove CF) seems likely to be greater than the difference between a peak Barry Bonds (elite LF, if such a thing exists) and a peak Alex Gordon (Gold Glove LF).

I know it's a different argument because of the nature of the positions, but if you moved Andrelton Simmons to first base, is there any possible way he could be so much more "above average" defensively that it would make up for his bat? I just don't think there are enough defensive opportunities at the position.
   71. bjhanke Posted: May 20, 2014 at 06:10 AM (#4709971)
My understanding, from experience, is that normal fans care about sabermetric stats just as long as they can understand them. This leads to two things - two kinds of "understanding": 1) A baseball fan who is good with numbers, especially algebra, or, double-especially, statistics, likes the new stats as long as he can follow them, and as long as his debating opponents in the local sports bars can't. 2) A baseball fan who is just downright bad at math likes new stats if he can understand what they are trying to do (measure wins, runs scored, runs allowed, etc.) and his sports bar opponents can't figure out how to do that any better than he can. In short, it's all about winning arguments in sports bars. When I was writing a weekly sabermetrics column about the Cardinals for the alternative weekly (The Riverfront Times here in STL), we got feedback that amounted to my column got tacked up to the corkboard in EVERY sports bar in town, just because it resulted in lots of drunk baseball fans arguing over whatever genetic mutation I'd trotted out that week, staying longer in the bar and drinking more. When the paper cancelled its entire sports department (having been bought by a chain based in Houston), we got told that it was just impossible to convince Houston newspaper owners of just how fanatic sports fans in STL were. We had me on baseball at the time that I was publishing books. Howard Balzer (who wrote the Sporting News Pro Football column) did that, and a good, though unknown, hockey guy could hold up his end of the deal. We caused more uproar in bars than the St. Louis Post-Dispatch guys did. But we could not convince the Houston guys that this was a possible thing; we just HAD to be making it up.

So the answer is yes, they do appreciate sabermetrics, but there are constraints on their comprehension, and for them, it all comes down weaponry for Happy Hour and the water cooler. If you write, go down to the bar incognito. It is truly an experience. - Brock Hanke
   72. CrosbyBird Posted: May 20, 2014 at 01:07 PM (#4710194)
It seems to me there are more non-routine opportunities in centerfield, and thus it would be possible for an excellent defensive outfielder to get higher above average at the position, even if the average is significantly higher.

That could be true, but I don't know that it is true. You mentioned Bonds and Gordon, but Bonds' best season defensively is in the mid-30s in run prevention and Gordon's is in the mid-20s. Our bar for a routine play for an average LF might be a lot lower than the routine plays expected from an average or even poor CF.

I mean, positional adjustments are definitely problematic. You've got sample size issues and a moving target all the time. (There was a time when 3B was a much more defensive position than it is today, for example.) On the other hand, WAR is still a pretty good crude model.
   73. cardsfanboy Posted: May 20, 2014 at 01:11 PM (#4710198)
For argument's sake, let's say that instead of Trout and Bourjos, the Angels had two identical Mike Trouts, one in CF and one in LF, with equal defensive ability and putting up identical offensive lines. Wouldn't the CF Trout have a higher WAR?

I am asking for enlightenment -- I actually don't know how it works.


At that point in time, it really depends on how they select to play defense. Again, an elite defender is made to look better by war, if he has lesser defenders around him. In this scenario, how often the LF Trout defers to the CF trout will be the determination of their ultimate value.

In theory, if you move an average centerfielder to left or right, he's going to up his rField enough to make up the positional adjustment. In reality, that is probably pretty unlikely, as the scenario that causes that to happen, usually means there is a better/elite centerfielder taking his place and that will ultimately hurt everyone else in the outfield.
   74. Ron J2 Posted: May 20, 2014 at 01:44 PM (#4710215)
#52 In another one of these threads I noted a little study I had done. I looked at players who had 300+ PAs in consecutive seasons and looked at the variation of WAR oWAR and dWAR. dWAR is the most stable of the three.

For the record

Players with 300+ PAs in consecutive seasons, 1955-2011 (4536 players). In case it's not clear, I'm comparing the same players in (say) 2011 and 2012.

Stat   Correlation  Standard deviation
WAR       .52             2.2
oWAR      .55             1.9
dWAR      .65             1.3
dWAR2     .61             1.3 


Note that there two sets of dWAR numbers. The earlier one uses TotalZone -- an adjusted range factor method since PBP based metrics were not available were not available until 2002. Year to year variation is about the same.

EDIT: That 1.3 win standard deviation is sufficiently high to be skeptical of the precision it's displayed to. I mean I understand why it's displayed to tenths of wins. But basically it's really best to think in terms of letter grades.

   75. Ron J2 Posted: May 20, 2014 at 01:47 PM (#4710216)
And #55 my best guess is that there's no meaningful signal on ~70% of all balls in play. They're either plays that everybody makes or balls that nobody gets to.
   76. puck Posted: May 20, 2014 at 02:54 PM (#4710260)
Going by their HOF results, I think that is about a true statement. Keith's best finish in the hof vote was 10.3 while Garvey's was 41.6...and this was in the mid 90's when people were making these votes.

HoF voters had Garvey's hits/RBI/HR/Batting avg stats, though.

They must not have had to sit through watching him try to throw a runner out a 2nd, though. Though his scooping ability was widely praised.
   77. Moeball Posted: May 20, 2014 at 04:24 PM (#4710316)
There is an argument that there is really a cap of runs a players defense can save over an average player of his position, when you see players putting up higher numbers than the theoretical cap, it's not because of otherwordly defense, it's because of weak teammates that they are able to compensate for.(or selective play....say a third baseman with fantastic range like Rolen in his prime, is constantly taking balls that the average shortstop fields, this is going to make Rolen look greater than a thirdbaseman who allows the shortstop to field the ball, and it's going to hurt Rolen's shortstop, because now he has fewer plays made, and it makes mistakes have a stronger impact on his numbers, even though at the team level, the results were exactly the same.) In some respects, defensive stats can be similar to rbi, in that they can be team dependent.


In theory, I guess this makes sense, but I keep thinking about the Orioles teams that had Brooks Robinson at third and Mark Belanger at short. You would think one player with spectacular range should be taking plays away from the other, yet they both grade out as not just Gold Glove level defense at their respective positions, but arguably the best ever to field their positions. Side by side all those years. How did that happen? Was it just that the Orioles pitchers were getting an absurdly high % of balls hit to the left side of the infield and somehow the "runs saved" numbers aren't adjusting properly for this?

They must not have had to sit through watching him try to throw a runner out a 2nd, though. Though his scooping ability was widely praised.


Garvey was actually an excellent fielder - at catching a baseball. When the Dodgers tried him at third in the early '70s he actually had decent range, but his throwing was horrible, so they moved him to first. Bill James has made much about the comparisons of Bill Buckner and Garvey as first basemen, noting that in the original award-giving aspect, Garvey was rated higher, then sabermetric measurements (Linear Weights, specifically) went the other way and said Buckner was better. The main difference was that when grounders were hit to Buckner the out always went 3-1 (pitcher covering first to take the throw from Buckner), whereas Garvey almost always recorded the out as 3 unassisted and would just run to the bag. Yes, Buckner received too much credit for assists on plays he might have been able to get himself by just running to the bag - but Bill is not taking into account that Garvey had to cover all those plays himself because even little short tosses to the pitcher were the sort of thing he could find a way to throw away. His arm truly was that bad.

On the other hand, to his credit - Garvey was great at picking it - both scooping low throws and jumping up to keep errant throws from getting away. Note, for example, that no one knew Steve Sax had a throwing problem until Garvey was no longer playing first for the Dodgers. I watched a lot of Dodger games in Sax' rookie season back in '82 and his throws were all over the place although he wasn't getting charged with a bunch of errors at the time. Garvey was just catching everything that came his way no matter where the throw went.

Now that I think of it, the total opposite to that was the situation in Chicago with the White Sox in the '90s. You know how a fairly common play seen on baseball highlights is the great diving play by the third baseman, who then has to make an off-balance throw to first, requiring the first baseman to also make a good catch on a wild throw?

I have tremendous respect for Robin Ventura when he was playing. He would not only have to make the diving play at third, but he would also have to make a good throw to first despite being off balance, because Frank Thomas couldn't catch a throw unless it was pretty accurate to start with. The digging out the throws in the dirt or snagging the high wild throws was not nearly in his repertoire the way it was with someone like Garvey.
   78. cardsfanboy Posted: May 20, 2014 at 04:42 PM (#4710327)
In theory, I guess this makes sense, but I keep thinking about the Orioles teams that had Brooks Robinson at third and Mark Belanger at short. You would think one player with spectacular range should be taking plays away from the other, yet they both grade out as not just Gold Glove level defense at their respective positions, but arguably the best ever to field their positions. Side by side all those years. How did that happen? Was it just that the Orioles pitchers were getting an absurdly high % of balls hit to the left side of the infield and somehow the "runs saved" numbers aren't adjusting properly for this?


Agreed, not sure what was happening there, but the point Bill James was making when talking about outfielders with absurdly high defensive totals,(per war) is that it does strain the level of credibility to think that three great outfielders could save 90 runs over the course of a 162 game season.

Yes, Buckner received too much credit for assists on plays he might have been able to get himself by just running to the bag - but Bill is not taking into account that Garvey had to cover all those plays himself because even little short tosses to the pitcher were the sort of thing he could find a way to throw away. His arm truly was that bad.


Bill specifically did mention Garvey's lack of an arm when discussing this issue, he used those two players as a point of contrast, not saying one or the other was the better defender,(although Garvey was the better fielder) but to point out the flaw of using a system that relied on assists for it's defensive evaluations.


I have tremendous respect for Robin Ventura when he was playing. He would not only have to make the diving play at third, but he would also have to make a good throw to first despite being off balance, because Frank Thomas couldn't catch a throw unless it was pretty accurate to start with. The digging out the throws in the dirt or snagging the high wild throws was not nearly in his repertoire the way it was with someone like Garvey.


I've mentioned it a couple of times, and I think that a good defensive first baseman is under valued by every defensive system out there right now. (I think that on average on a per game basis, the pitcher, catcher and first baseman are more valuable players than other positions) Most(if not all) defensive systems more or less look at whether the play was made or not, and doesn't really care what impact the receiver had on the play. I think eventually we'll get to a point in analysis where first baseman are credited rightly so for good defensive reception. It shouldn't be a system set up as "relative to average" though, it should be a bonus taken from the defender that is being given credit for the play made.

For example, if a second baseman fields a ball and records an out, and that particular play is worth .3 runs above average, but he made an errant throw that the first baseman saved, the second baseman gets .2 runs above average credit, while the first baseman gets .1 bonus run saved credit. (all numbers are theoretical for illustrative purposes only)
   79. Johnny Sycophant-Laden Fora Posted: May 20, 2014 at 06:21 PM (#4710361)
et they both grade out as not just Gold Glove level defense at their respective positions, but arguably the best ever to field their positions. Side by side all those years. How did that happen?


From 1968-1973 the Orioles had a team ERA of 2.87 and a FIP of 3.34
During that span league FIP was about 8-10 points higher than league ERA, so let's say the Orioles would have had an ERA of 3.24 with an average defense, 3.24 to 2.87, 0.4 runs per game or 65 runs a year

But to answer your question- statistically Belanger's and Brooks' defensive primes largely do not line up - Belanger's number went up when Brooks' went into decline

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
The Id of SugarBear Blanks
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOT: Politics - December 2014: Baseball & Politics Collide in New Thriller
(4537 - 7:30am, Dec 18)
Last: Omineca Greg

NewsblogOT: Soccer December 2014
(295 - 7:16am, Dec 18)
Last: ursus arctos

NewsblogThe 2015 HOF Ballot Collecting Gizmo!
(28 - 6:59am, Dec 18)
Last: The Duke

NewsblogAZCentral: Miley's Preparation Apparently an Issue for DBacks
(15 - 6:39am, Dec 18)
Last: villageidiom

NewsblogMLBTR: Padres-Rays-Nationals Agree to Three-Team Trade
(35 - 5:10am, Dec 18)
Last: Bring Me the Head of Alfredo Griffin (Vlad)

NewsblogRoyals sign Edinson Volquez for two years, $20 million
(4 - 4:58am, Dec 18)
Last: Bring Me the Head of Alfredo Griffin (Vlad)

NewsblogOT: Monthly NBA Thread - December 2014
(643 - 2:32am, Dec 18)
Last: Howling John Shade

NewsblogMorosi - Effects of US Shift on Cuba Policy
(8 - 1:46am, Dec 18)
Last: stevegamer

NewsblogAre Wil Myers' flaws fixable? | FOX Sports
(74 - 1:04am, Dec 18)
Last: PreservedFish

NewsblogThe Dan Shaughnessy Hall Of Fame Ballot
(58 - 12:17am, Dec 18)
Last: SoSHially Unacceptable

Hall of MeritMost Meritorious Player: 1960 Ballot
(6 - 12:08am, Dec 18)
Last: lieiam

NewsblogSource: Myers to Padres in 11-player deal with Rays, Nats | MLB.com
(3 - 11:39pm, Dec 17)
Last: zonk

NewsblogOT: NFL/NHL thread
(9148 - 10:39pm, Dec 17)
Last: Pops Freshenmeyer

NewsblogOrioles agree to one-year deal with LHP Wesley Wright, pending physical, source says
(11 - 8:58pm, Dec 17)
Last: escabeche

NewsblogIndians sign Gavin Floyd to deal
(9 - 8:37pm, Dec 17)
Last: Dock Ellis on Acid

Page rendered in 0.8914 seconds
48 querie(s) executed