WAR Graphs
If you’re on twitter and you follow some of the same people I do, you’ve probably been inundated with one thing over the last few days – WAR Graphs. Perhaps one of the most underrated parts of the site, it is especially fun around times like this when Hall Of Fame voting is a popular topic of discussion. Mike Axisa was having so much fun using the graphs to compare different players that he opened a twitter account, @WARGraphs, specifically for the purpose of sending out such gems as the one below. Even though it has nothing to do with Cooperstown, this one was one of my favorites.
It’s funny how our memories force us to forget that, for a time, Raul Mondesi was a pretty terrific player. It is great how a simple chart can offer an interesting reminder.
There’s also other uses for the WAR graphs, of course. Because Patrick Sullivan made a reference to Jim Rice the other day, I found this one that is simultaneously hilarious, interesting, and kind of sad.
Given how much interest there is in sharing some of these on twitter, I thought it was worth reminding regular readers about a portion of the site we don’t talk about enough. WAR Graphs, one of the more fun tools around around, and a great way to kill a few minutes each day.
Dave is the Managing Editor of FanGraphs.
I love this part of the site but never use it! It’s too much fun!
I’ve never looked at that part of the site before…it’s really cool. Is there any way to make a graph for pitchers?
Where is Curt Schilling? Are some guys just not showing up on this or am I completely whacked.
Just poking around, but this one was fun…
http://www.fangraphs.com/graphsw.aspx?playerid2=335&playerid3=826&playerid4=&playerid5=
http://www.fangraphs.com/graphs/1011066___1009904_sgraph_%20_12_29_2010.png
Fun Raul Mondesi fact: Raul once asked the Blue Jays assistant general manager for a mid-season loan to purchase a fire truck. No word on whether Given could come through with the loan or not.
Okay, I may just be whacked, but Kevin Towers still has a sour taste in my mouth. Haven’t liked one thing. Not one move so far, and very little has been written about it.
But anyways, it looks like this is only for hitters. Besides following this link is there a way to compare pitchers?
Nope. Pitchers are an urban myth.
I think what the Mondesi-Braun graph really shows is …
[1] The difference in perception of offense and defense.
[2] We might not be calculating/attributing defensive runs correctly.
Mondesi gains up to 2 WAR/y on Braun in defense some years, and some years Mondesi’s defensive runs vary by 20 RUNS!!! Seriously in ’02 he was -16 runs and in ’03 he was +5. Are we talking about the same player here?
One of the things that bothers me about WAR is that it doesn’t necessarily represent what actually happened (wOBA, DRuns, FIP, etc), but what should have happened on the average occasion of that event. At least that seems to be the case to me.
I really think we should not put so much emphasis in the WAR comparisons until we’re more confident in our defensive metrics, and I don;t think at this point we should be all that confident (single season).
Sheesh, Modesi was either below replacement level or at/above league average … depending on year-to-year variations in defensive runs. Looking at his defensive run value at FG, I have no idea whether Mondesi was a good or terrible outfielder. My observations and memory tells me he was adequate fielder with an absolute laser. But for his career (13 years) he has negative defensive runs 7 times and positive runs 6 times.
I’d prefer NOT to use that metric as a comparison to Ryan Braun. Braun is pretty much a consistently below average fielder and absolute masher at the plate with decent baserunning numbers.
It’s unfortunate, because IMHO, I truly believe that WAR is one metric that EVERYONE can grasp and use … which will lead to a whole new appreciation for pitcher’s skill, defensive value, the luck aspect, etc …. but until we get stuff like this ironed out, WAR is just going to be another “hammer” that people use for all instances (i.e., when all you have is a hammer, everything is a nail).
You are really, really misusing that hammer/nail saying.
As for Mondesi, his reputation was as a physically gifted player prone to bad decision-making, whose bad attitudes sometimes bled over into the field. Does it really kill UZR that a player notorious for not consistently playing to his abilities was an inconsistent defender?
WAR is only tremendously valuable is the defensive part is accurate. I say tremendously valuable based on how we use WAR … to evaluate EVERYTHING.
If it were just a fun stat, it would be different. But we use it to evaluate contracts, GMs, trades, seasons, awards, careers, etc. It literally is our hammer.
It’s important to me for this reason, and also because I think the average fan can grasp and use WAR, which may lead to increased knwoledge and/or appreciation of some under-valued aspects of the game.
It’s very difficult for me to attribute a difference of 2 WAR in defensive value to attentiveness, decision-making, and mood when looking at a single player in consecutive years. I looked up the DR of Rickey Henderson and BJ Upton, and there’s not the extreme +/- of Mondesi.
So, what I am really getting at is ….
[1] What is the range of randomness/variability of DRuns? (My concern is it might be in the 10-15 runs range, which is HUGE, specifically in regards to its potential affect on player WAR).
[2] Would we be better off stating a player’s WAR as a possible range? (Then the potential problem is that rnage is so great, 2.5 – 3.5 WAR) that it becomes too general to be tremendously useful.
A 15-run swing is basically around 25 plays missed or made from one season to the next. That is seriously not that big a difference. You didn’t bat an eye at Mondesi’s offense varying far more than that. No, a non-pitcher does not have as many chances to impact the score on defense as he does on offense, but it’s not as unequal as you’d think. Mondesi was having hundreds of balls his his direction every year, is it really surprising that he could field one more a week one year than he did the last?
The reason 15 runs seems like a big difference is that we think of the difference between a gifted and a hopeless defender as only about 20 runs, which it probably is if we’re talking true talent. A big part of that is that the defensive spectrum has separated out the truly hopeless defenders and moved those who aren’t all-world to lesser defensive positions.
That’s why I am asking. Because a 15-run swing, based on how we would WAR, could be the difference of ~ $7M in on-field value.
To me, that’s a big deal.
That’s basically the strugle I have with trying to picture how it would look on the field. One year a guy could have the value of Jim Edmonds and the next the value of Chris Duncan, and I cannot wrap my mind around that idea, knowing the visual difference between the 2 players (from my frequent observations).
I would think that defense would be a consistent component of player’s performance, as oppossed to the offensive runs which could be heavily dependent on BABIP.
It’s possible I am missing something, and that’s why I am asking. If the methods of defensive runs could swing a player’s value 1+ WAR in either direction, that’s a problem to me … based on how we use WAR, and the potential importance of WAR. By that I mean, WAR has the potential to change everything in how fans view players, teams view players, how GG’s are awarded, etc.
Why should defense be more consistent? I think it should vary more do to a players health than hitting. A sore knee should slow a guys ability to run and dive far more than his ability to hit a home run (Kirk Gibson), strike out, or take a walk.
Oh my god, you with this shtick again.
Put aside all of your concerns about WAR, what else would you use to value those two players?
Oh relax. I’m not asking for an alternative to WAR. I am/was asking a very specific question of defensive runs.
What I noticed is that Mondesi’s value depended a lot of defensive runs, and I could not figure out how someone could go from elite to below average while his speed remained (i.e., stolen bases), his arm was always strong, and he remained with the same team in the same stadium.
Looking into some more, after ’97 teams stopped running on him (OF AST cut in half), as well as splitting time in CF/RF. But, that’s not it b/c he had 19 OF AST in ’01, and put up – DRuns.
So, I’m left to think it has to do with his “range”, and I’m back to the original question.
Found this from Litchman …
Note to self: Look for answer first, then ask the question if the answer cannot be found.
How many times are you going to make this post? Good lord.
While it may be a bit overplayed for the defensive stuff, what’s the point of hating on discussion? WAR is used a ton and could use some significant improvements.
With that said, in my opinion the discussion should generally be about what can be done to improve on it at this point- the inadequacies of Fangraphs WAR should be pretty well noted by this point. Rather than having people restate them all the time, maybe we should just keep a list of them somewhere on the site so people can refer to “WAR defense issues, point 5” rather than having to restate it?
That would also give us a good idea of what really needs improving upon, versus what merely appears to be an issue but actually is a fundamental issue of measuring chaotic/random processes.
Why is it difficult to believe that players can have exceptional and poor defensive seasons? It seems to be standard operating procedure for critics of UZR to argue that defensive performance from 1 year to the next should be consistent. However, no one is surprised when a hitter performs terribly at the plate 1 season or when a pitcher performs horribly in 1 season. Why is it difficult to believe that Mondesi might have been exceptional defensively 1 year and slightly below average the next? It happens to pitchers and hitters all the time and no one says a thing!
Simply put, defense is a more individual skill/act, than getting a hit.
Far fewer factors influencing.
I would have thought that was obvious.
I actually disagree with that, due to injury issues. For most positions, mobility is key. For any player with a good bat, they’ll play you if you can take the field. Hence, one should expect a good batter to have some outlier seasons where their defensive value is hindered by lingering injuries.
Basically, I’m of the opinion that there could very well be significant shifts even in “true talent” (over that season) with respect to defense.
Chet Lemon v Jim Rice …
Quite a bit of his WAR value is defense, and while being more consistent than Mondesi, it still goes to extremes.
Heck even Jim Rice bounces around from +10 runs to -5 runs to ….
What is the randomness or error margins for defensive runs? It seems like it could be VERY large, so large as to dramatically skew the overall player value.
But, at least, the latter comparison shows the importance of walks and defense in player value.
The entire defense component of WAR is problematic. Look inside UZR and you’ll see plenty of instances of +/- 10 run differences for home/road. Arms rated vastly differently home/road. Or both errors and team double plays as components. Neither of those only measure the ability of one player, and the range component is obviously so subjective as to be not credible. Don’t get me wrong, I reference it, and I think it’s useful, but UZR as a component of an all-encompassing measure of player performance is just silly in my opinion.
While I appreciate your drumbeat on the issue, Circle, I think there is a bias at work where the current formula makes it a convenient tool to attack scouting conventional wisdom about both historical and current players. Unfortunately, for that reason I don’t really expect any changes.
I looked up one of my personal favorites, Willie McGee, who is as absent-minded & talnted as a defender could be. On consecutive plays he could (literally) take away a triple by catching a ball at the track (on the run) and then on the very next play break back on a ball 4 steps, and then realize it’s a blooper and sprint in only to have the ball land, bounce over his head, for a bases-clearing “triple”.
His defensive runs all all over the place in terms of + or -, but consistently withint the range of -8 to +8. That can play a significant role in his yearly WAR.
His batting runs are also vary quite a bit, but one could likely attribute that to BABIP (ranges from .270 to .395).
Cool feature.
I just checked out:
[1] Ryne Sandberg
[2] Roberto Alomar
[3] Bobby Grich
[4] Lou Whitaker
The cool part is not looking at the key and trying to guess what linebelongs to each player.
How is Whitaker not in the HOF? I know there have been some bad choices lately but check this out:
http://www.fangraphs.com/graphsw.aspx?playerid2=1011411&playerid3=1013846&playerid4=1010897&playerid5=1003091
because he did not play in NY, Boston, or Chicago.
You mean like Ron Santo?
Yet another reason to bring up my Biggio-vs-Whitaker rant:
http://www.fangraphs.com/graphsw.aspx?playerid2=549&playerid3=1013846&playerid4=&playerid5=
Whitaker lacks the really big seasons and fares poorly comparatively in his top 6 or so. Trammell, Grich, Sandberg, and Biggio each have at some big seasons. Larkin and Alomar have a distinct edge in the top half-dozen ones.
If peak value is heavily weighted by voters, you could see why Whitaker didn’t receive much support. To me, he is borderline, though he definitely should have stayed on the ballot long enough to receive some extended consideration.
Peak > than consistently very good (in the mind’s of the writers).
Sandberg had a much greater peak. Whitaker was better for long, or in other words, held and even increased his value as he aged.
I compared Morgan, Grich, Whitaker, Sandberg. Ryne and Joe have the peaks, the other two sustained.
Why all the love for Whitaker when Grich outperformed him?
Obviously, there’s an underappreciation of 2B and 3B.
Maybe it’s just me but I find the WAR graphs thing to be a nice tool, but a pain in the ass to use. Unless I’m missing some special way to type in player names, I’m just going to look at other’s graphs.
Type in the last name (or just a part of it, and it zooms you right to that section of the list). For God sakes, do not scroll down the entire list to get to “Williams”.
Type the last name in the box. Compare 4 players in 15 seconds.
Done that several times. Maybe it’s my browser?
Did it several times more, just to try it again. I’m using Internet Explorer 9 Beta version.
I’ve said it before but these graphs can illustrate it nicely; Bobby Abreu is really underrated. Try running him against guys like Ichiro or Canseco or Miguel Cabrera or Gary Sheffield, Abreu stacks up pretty well. His skills, grade A OBP and A+ durability, aren’t all that sexy but they sure produce runs.
I had been trying to get to the WAR graphs feature from player pages via Compare, and couldn’t figure out what I was doing wrong… I was trying to compare Andre Dawson and Dwight Evans, and now I can.
Brooks Robinson vs Ron Santo
Try Pujols, Mays and Aaron…its actually pretty interesting.
Go on already, go check it out.
Just for fun, add A-Rod to the mix. All by way of saying that Pujols, especially with his injury history, is unlikely to keep up Aaron’s pace into his 40’s. Wait, am I stating the obvious? Wasting virtual ink? Praising by faint damnation? Definitely one of those.
I think its unlikely that ANY player will ever keep up Aaron’s pace like that. He was unique.
A-Rod was interesting too.
Welcome to my personal blog!
Classy anonymous jab! You know, people are allowed to comment at this site. It’s sort of what the comments section is for.
Ichiro and Tim Raines is an interesting one, with the HOF debate in mind.
http://www.fangraphs.com/graphsw.aspx?playerid2=1011586&playerid3=97&playerid4=1011055&playerid5=1001124
4 Hall of Famers right here
Easily. Considering Michael Jack was one of the Top 10 position players of all time probably.
Brooks and Bogg are also no question and Larry (pains me to say it as a Phillies fan) has had one hell of a career. Well done Larry, well done.
Zimmerman, if he stays healthy, could have a career similar to Michael Jack’s. Great defense and a nice power bat. Sure, he’ll likely never be a 12 Win player but he could end up fitting in on that graph too.
That should say “10 Win player” not 12. Typo on my part about Michael Jack.
Jeter vs. Trammell
That is more tragic than it is surprising.
Trammell was 9th best shortstop of all-time according the the Bill James Historical Baseball Abstract.
It’s mystifying that he has still only garnered less than 1/3 of the votes he needs for enshrinement.
I was a huge Tram fan but he’s a Hall of Very Good guy. He just couldn’t stay healthy in the 2nd half of his career. It a shame too as he was spectacular during the 1st half (see also, MATTINGLY, Don)
That doesnt take anything away from Tram though. It is Jeter’s good luck to stay healthy and productive well into his 30s.
UZR is best when looked at over 3 year periods. It’s certainly not perfect, but it’s the best we’ve got.
Ichiro, Brett Butler, Mike Cameron, Tony Phillips.
Four of my favorite players to watch; they mesh together pretty closely.
Also fun: Bobby Grich and Lou Whitaker schooling Ryne Sandberg and Jeff Kent.
Chet Lemon, Hall of Fame ?
I guess that would create a ‘sad’ feeling about a measure of player value missing the mark like that.
Great comparison though, both came up in 1975, played OF in AL East and finished careers at about same time. Call me old-school, pulled up their Baseball-Reference pages: It looks like Chet’s defense was solid but I don’t see any hardware to back it up; and Jim Rice had some beastly years. I dig the similarity scores on that one too (Orlando Cepeda, Andres Galaraga, Duke Snider, Ellis Burks, Joe Carter, Vlad Guerrero…).
And lay off the Circle Change, he’s having a good time. Time which he’s got plenty on his hands for winter break.
Can anybody provide a quick and dirty guide to reading these graphs?
Actually, I just need to know what nth season means…
nth best season means that the seasons are ranked in order from best to worst.
1 (actually it’s .1 on the graph) is the players’ best season
2 (.2) is the players’ 2nd best season
etc…
Thanks guy.
will this ever be available for pitchers?
Nth best season is just sorted by best season, 2nd best, 3rd best, etc.
Why is there no obvious menu item for WAR graphs? Seems you have to link to it from an article.
This.
Why is there no tab for it?
There is a (sub)tab for WAR Graphs; at the top of the page, Leaders ->WAR Graphs. There you go.
[1] Lou Brock
[2] Ichiro
[3] Tim Raines
[4] Kenny Lofton
Lofton may one day be the ‘Ron Santo’ of baseball, only without the sympathy or reverence.
A major component of respect in the game, aside from the ability to hit homers, is not being a complete and total a-hole! As it turns out, Lofton was much better than I knew at the time.
Just for grins, I think I’ll try Albert Belle!
http://www.fangraphs.com/graphsw.aspx?playerid2=1010978&playerid3=1010977&playerid4=&playerid5=
http://www.fangraphs.com/graphsw.aspx?playerid2=1001918&playerid3=1001919&playerid4=&playerid5=
baseball success is not all genetics. They were twns
leave billy alone! He had a very good glove
I laughed.
http://www.fangraphs.com/graphsw.aspx?playerid2=1109&playerid3=1001157&playerid4=327&playerid5=1005044
It’s a shame how long that Ken Griffey, Jr. line was basically flat.
He was still one of the top CFs in history though. He had one hell of a run in the 1990s.
http://www.fangraphs.com/graphsw.aspx?playerid2=547&playerid3=1266&playerid4=1086&playerid5=455
Bagwell doesn’t really belong in this grouping with Edgar, Palmeiro, and Walker.
I’m surprised wargraphs doesn’t like Schmidt a bit more. Was his defense not as exceptional (can play ss!) as reported at the time?
How are the older defensive scores even being calculated? Schmidt’s scores are pretty good, but looking at a guy like Brooks Robinson, who put up +32 and then +33 back to back (and granted, a guy who is widely regarded as one of the greatest fielders ever and the greatest fielding 3B ever), shows some pretty insane numbers.
Upset that I didn’t know about this feature before now.
http://www.fangraphs.com/graphsw.aspx?playerid2=1013157&playerid3=335&playerid4=1012186&playerid5=826
Basically the same WAR and the only one who won’t be a Hall of Famer is Trammell. Stupid!
Tram should have stayed healthy after Age 30. Sure, its complete BS but he’ll forever be in the Hall of Very Good as a result of that. Though I wouldn’t be shocked if he eventually gets in as a Veteran’s Committee selection in 10-20 years.
That’s just how it is.
http://www.fangraphs.com/graphsw.aspx?playerid2=1201&playerid3=1003865&playerid4=&playerid5=
I loved this one.
http://www.fangraphs.com/graphsw.aspx?playerid2=1003865&playerid3=1003864&playerid4=&playerid5=
Too bad we don’t have a WAR for Bob Evans.
http://www.fangraphs.com/graphsw.aspx?playerid2=970&playerid3=1012186&playerid4=1153&playerid5=96
I love the comparison of Edmonds to Ozzie Smith.
Wow! Pujols, Gehrig, Aaron, and Musial!
http://www.fangraphs.com/graphsw.aspx?playerid2=1177&playerid3=1004598&playerid4=1000001&playerid5=1009405
Gehrig’s truly spectacular years really stick out there. I mean, 3 seasons well over 10 Wins…just insane.
Look what happens when you swap out Musial with Jimmie Foxx
http://www.fangraphs.com/graphsw.aspx?playerid2=1177&playerid3=1004598&playerid4=1000001&playerid5=1004285
Foxx is the only one that comes close to matching Gehrig’s best year ironically. Though he faded fast (mostly due to being an alcoholic IIRC).
This is pretty interesting, look at the WAR by age for these 3Bs
http://www.fangraphs.com/graphsw.aspx?playerid2=1011447&playerid3=1011586&playerid4=97&playerid5=1001400
http://www.fangraphs.com/graphsw.aspx?playerid2=1002841&playerid3=1010897&playerid4=1001458&playerid5=
Jose Cruz, Jim Rice, and Lou Brock.
http://www.fangraphs.com/graphsw.aspx?playerid2=1000453&playerid3=1000137&playerid4=1000799&playerid5=1000505
Frank “Home Run” Baker, Dick Allen, Buddy Bell, and Sal Bando.
http://www.fangraphs.com/graphsw.aspx?playerid2=1011586&playerid3=1008236&playerid4=1274&playerid5=1001124
Alex Rodriguez, Mike Schmidt, Eddie Mathews, and Wade Boggs. What’s this going to look like in 2-3 years?
Doesn’t really matter. Schmidt will still be the best 3B in history.
To be fair, A-Rod’s most, if not all, of A-Rod’s top seasons came at the SS position. Had he put those numbers up as a 3B, his WAR during those seasons would drop. So you’re probably right.
http://www.fangraphs.com/graphsw.aspx?playerid2=1119&playerid3=1008675&playerid4=1005692&playerid5=
Jeff Kent, Bid McPhee, and Billy Herman.
http://www.fangraphs.com/graphsw.aspx?playerid2=1002103&playerid3=1002318&playerid4=1005706&playerid5=1002043
Will Clark, Orlando Cepeda, Keith Hernandez, and Norm Cash.
http://www.fangraphs.com/graphsw.aspx?playerid2=1013133&playerid3=1003271&playerid4=1003881&playerid5=1002384
Joe Torre, Bill Dickey, Buck Ewing, and Mickey Cochrane.
http://www.fangraphs.com/graphsw.aspx?playerid2=1275&playerid3=1004101&playerid4=1002015&playerid5=1000898
Pudge Rodriguez, Carlton Fisk, Gary Carter, and Yogi Berra. Who’s the second-best C of all-time, behind Johnny Bench?
Yeah, this thing is pretty fun, huh?
http://www.fangraphs.com/graphsw.aspx?playerid2=1003865&playerid3=1010897&playerid4=1007872&playerid5=
Dwight Evans, Fred Lynn, Jim Rice.
Yet the first two never even got a whisper of HoF support while the latter has the more ridiculously organized campaign in the history of baseball (outside of maybe Rizzuto perhaps).
How the hell did Rice get in but Evans never get more than 10% of the vote. At least Lynn and Evans were great defenders during their career.
Also, Dick Allen was a much better hitter than Jim Rice and he’ll never get in either…life isn’t fair.
Doing a graph with Chase Utley, Ryne Sandberg, Craig Biggio and Robert Alomar, the story isn’t how Utley belongs in that group, peak-wise, but that all hit a huge wall at the same age, 34:
http://www.fangraphs.com/graphsw.aspx?playerid2=1679&playerid3=860&playerid4=549&playerid5=1011411&wg=3
If Utley hits that same wall at 34, he doesn’t go HOF with his late start; if he can soften his landing some, then I think he has a chance.