AngelsWin.com Fan Forum
AngelsWin.com Fan Forum
Search | Home | Chat | Blog | Store | Interviews | History Forums | Quotes | Language
You are logged in as a guest. ( logon | register )



Random quote: "In the 7th inning, the fans get up and sing 'Take Me Out to the Ball Game', and you're already there. It's really a stupid thing to say"--Larry Anderson
- (Added by: socaliente)


ESPN: Let's corral the WAR horse
Moderators: tomsred, Adam, IEBRUIN, Bruce Nye, Kurt Swanson, Blarg, mancini79

Jump to page : 1
Now viewing page 1 [50 messages per page]
View previous thread :: View next thread
   AngelsWin -> Los Angeles Angels DailyMessage format
 
gotbeer
Posted 2013-02-01 8:11 AM (#1931877)
Subject: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 20685

Let's corral the WAR horse
Top of the page Bottom of the page
Soto Speaks
Posted 2013-02-01 8:33 AM (#1931886 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse




Location: Irvine
Don't know what's funnier, the comments or the avatars of the people making the comments.

Like it or not, WAR is now a big part of the baseball conversation when comparing players.

Top of the page Bottom of the page
wopphil
Posted 2013-02-01 8:39 AM (#1931890 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse


All-Star

Posts: 3435

WAR seems like a good stat for evaluating hitting performance, but terrible for evaluating pitching. Guys like Weaver don't have very strong WARs because they don't strike out a ton of guys, despite logging lots of innings and putting up very good ERA (the single most important pitching stat, in my opinion). Seriously, how does Max Scherzer put up a better WAR than Weaver when Weaver had a far better WHIP, ERA, batting average against, more wins (I know, a useless stat, but Weaver was nonetheless superior), more innings pitched, and fewer walks? Park factor can't account for all of that.
Top of the page Bottom of the page
HaloMagic
Posted 2013-02-01 8:48 AM (#1931895 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 5418

I agree that WAR isn't the end all. But it's one of the only stats (maybe the only stat) that you could actually use in place of at least 10 or so individual stats and still get an accurate picture of a player from. With many stats you still need a great deal of other stats to adequately judge a players worth. You can't use BA and decide how much a player helped a team. Using HR and RBI doesn't give you a great picture of a player's contribution. It's great because it is essentially a rate stat (like OBP or SLG) but has been altered to work like a counting stat (like HR or runs). You don't have to try to guess the impact he made without knowing info like PA. You don't have to see a stat like 10 HR and not be impressed just because you didn't know it was over only 65 at bats. It allows you to know how much contribution a player should have added with his play over the amount of time he actually player. It might not be as good as a stat like WPA in determining actual contributions but it is much more predictive and avoids the problem of judging a player's worth based on good or bad luck.

As a fan, I obviously can't deal with just one stat. Like many or probably most on here, even before creators of these advanced stats were born I would dive as deep as I could into stats. That started as a child looking at the numbers on the back of my baseball card and being amazed at how easily I could compare the value of players that I didn't even really know. BA, RBI, and R are great but I definitely enjoyed when OBP and SLG came along and considered them welcome additions. I never stopped looking at a particular stat just because another one came along. Even now when looking to gauge a player I know that RBI and R are not a good way to determine worth. But I still look at them. I don't just skip past them. I've been looking at these numbers for so long it's impossible at this point to just simply ignore them. For me more numbers is always better.

But for a new fan of the sport just learning it and trying to understand the game I wonder if the more numbers the better mentality still holds true. I doubt it. More numbers would be confusing. They wouldn't let you really see the value a player has. For me I know RBI are worthless as a predictive stat but since I've been learning them for so long I can't just ignore them. But for a new fan, why would they want to use RBI to judge value? The stat isn't helpful and it seems like it would be mostly confusing to them to try to explain why we still look at it. Because there is no real answer. We just do. But with WAR they don't have to. They can look only at what provides value. It's definitely a little harder to calculate but to that I'd say two things: First, it's not impossible. The calculations are still arithmetic, not calculus. Second, you don't need to do the math anyway. How many sites show a players WAR now? Basically every stat site out there. Newspapers are carrying WAR now. Certainly some of the fun of stats as a kid was calculating them but that wasn't necessary for my enjoyment of the game.

So I say WAR isn't overused. Not more than any other stat. Many people ignore entire sections of stats if a player has a high OBP or hit a lot of home runs. They'll quote only the stats that help their point. This isn't new. But in my opinion, more so than any other stat, WAR is very close to equal with all of it's individual components. You could take all the single numbers that make up WAR and have only a slightly better picture of a player than you'd have by just WAR itself. That's something most other composite stats can't say.

If I could make some changes though, I'd say: A) Stop using pitching WAR. It's almost completely useless if it's from BR and if it's from Fangraphs it is completely useless. And B) I'd like people to breakdown oWAR and dWAR. dWAR at this point is still a little hard to trust and I'd like to know right off the bat if a player's value is tied mostly to their bat or their glove. If it's tied to their glove it tells me I should do a little more digging to get a more accurate picture of the player. The given WAR may still match my view but I know it's far more likely to do so if the value is tied to the bat versus the glove. I'd like to know how much I should trust the number and I feel separating those two out gives me a better idea of that.

Edited by HaloMagic 2013-02-01 8:54 AM
Top of the page Bottom of the page
jshep
Posted 2013-02-01 9:15 AM (#1931905 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse


Hall of Fame

Posts: 10189

Another Lackey thread?

Slow down, guys.
Top of the page Bottom of the page
Angelsjunky
Posted 2013-02-01 9:39 AM (#1931919 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 7388

Location: Limbo
I agree that WAR doesn't work so well for pitchers. It also doesn't work so well for comparing pitchers with hitters in terms of overall value. For instance, using wopphil's example of Jered Weaver who had a 3.0 fWAR in 2012 - let's look at what fWAR reveals. Remember, Jered Weaver had a 2.81 ERA in 188.2 IP.

- Weaver was less than half as valuable as Felix Hernandez (6.1) and Justin Verlander (6.8)
- # of pitchers with higher fWAR than Jered Weaver: 34
- A sampling of pitchers with higher fWAR than Weaver (2.81 ERA in 188.2 IP): Ian Kennedy (4.02 ERA), Jon Lester (4.82 ERA), Jeff Samardzija (3.82 ERA), etc
- # of position players with high fWAR than Weaver: 71
- A sampling of players with higher fWAR than Weaver: Omar Infante (3.2), Neil Walker (3.3), Ben Revere (3.4), Mike Moustakas (3.5, and a .242/.296/.412 line!), etc

Some questions that arise:
1) Who would you rather have in your rotation, Felix and a replacement level starter or two Jered Weavers? According to fWAR the two groups are of equal value.
2) Who was more valuable to their respective teams last year, Jered Weaver or Mike Moutakas (or any number of players)?
3) Was Jered Weaver really the 35th best pitcher in baseball last year?
4) Was Jered Weaver really the 106th best player overall in baseball last year?
5) When will the madness end?!

I like WAR, I use it a lot - but something is wrong here.
Top of the page Bottom of the page
Floplag
Posted 2013-02-01 9:43 AM (#1931923 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



All-Star

Posts: 4934

It isnt great, but its better than nothing.. is where WAR sorta falls.
It fails when you see guys like Bourne so high above guys that clearly they should not be...
the formula needs some slight tweaks, but again, better than trying to interpret multiple stats.
Top of the page Bottom of the page
Angelsjunky
Posted 2013-02-01 9:59 AM (#1931932 - in reply to #1931923)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 7388

Location: Limbo

"I kick ass at WAR."

Top of the page Bottom of the page
Mile High Dreams
Posted 2013-02-01 10:11 AM (#1931935 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 7978


War... Huh... Yeah!
What is it good for?
Absolutely nothing!
Uhuh... uhuh...!

War... Huh... Yeah!
What it is good for?
Absolutely nothing!
Say it again y'all
War... Huh... Look out!
What is it good for?
Absolutely nothing!
Listen to me - AAH!
War I despise

Top of the page Bottom of the page
AngelsOwnAll
Posted 2013-02-01 10:19 AM (#1931939 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 10511

Location: The part of the OC that the OC wishes was LAC
F WAR.




Top of the page Bottom of the page
HaloFan85
Posted 2013-02-01 11:57 AM (#1932002 - in reply to #1931886)
Subject: Re: ESPN: Let's corral the WAR horse




Soto Speaks - 2013-02-01 8:33 AM

Don't know what's funnier, the comments or the avatars of the people making the comments.

What's even funnier is the actual number of comments.  Ever since they made you log into facebook to post comments, the numbers have gone from 5000+ to less than 100 in most articles.

Top of the page Bottom of the page
Halo
Posted 2013-02-01 12:43 PM (#1932048 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



"My issue is this: I don't like the increasing over-use of (and over-reliance on) WAR as THE definitive evaluation of a player's worth.

This was particularly true during the Mike Trout-Miguel Cabrera MVP debate last fall. For instance, consider this headline on an ESPNLosAngeles.com story in late September: "Mike Trout Is Your MVP (WAR Says So)."


Headlines are meant to grab attention I don't know about that article, but you can easily make a case for Mike Trout without using WAR... and many did just that. Like other statistics that try to measure a player's value (RC27, VORP), WAR is a combination of many statistics that everyone uses. Simply looking at all the numbers... including things like double plays, and extra bases taken... should make people understand that Trout's value, offensively, is close to Cabrera's, at worst. Throw in defense and it is a no brainer. If nobody had created "WAR", that'd be the case.

It's not like there weren't people that weren't on the "Cabrera should be MVP becuase Triple Crown says so" bandwagon... or "Cabrera should be MVP because his team played in a weak division".
Top of the page Bottom of the page
dochalo
Posted 2013-02-01 12:59 PM (#1932066 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse


Hall of Fame

Posts: 7484

WAR is like the Target or Walmart of stats. A decent all-in-one with limits.
Top of the page Bottom of the page
Brian Ilten
Posted 2013-02-01 1:05 PM (#1932067 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 7680

Location: Sec 526 Angel Stadium
Make Love not WAR?








Top of the page Bottom of the page
AngelsLakersFan
Posted 2013-02-01 7:12 PM (#1932405 - in reply to #1931890)
Subject: Re: ESPN: Let's corral the WAR horse



All-Star

Posts: 4978

wopphil - 2013-02-01 8:39 AM

WAR seems like a good stat for evaluating hitting performance, but terrible for evaluating pitching. Guys like Weaver don't have very strong WARs because they don't strike out a ton of guys, despite logging lots of innings and putting up very good ERA (the single most important pitching stat, in my opinion). Seriously, how does Max Scherzer put up a better WAR than Weaver when Weaver had a far better WHIP, ERA, batting average against, more wins (I know, a useless stat, but Weaver was nonetheless superior), more innings pitched, and fewer walks? Park factor can't account for all of that.


Angel stadium was the best pitching park in baseball last year (I believe) and Detroit actually played towards offense. The Tigers defense was also among the worst in the AL. Angels defense was stellar.
Top of the page Bottom of the page
AngelsLakersFan
Posted 2013-02-01 7:47 PM (#1932425 - in reply to #1931895)
Subject: Re: ESPN: Let's corral the WAR horse



All-Star

Posts: 4978

HaloMagic - 2013-02-01 8:48 AM

I agree that WAR isn't the end all. But it's one of the only stats (maybe the only stat) that you could actually use in place of at least 10 or so individual stats and still get an accurate picture of a player from. With many stats you still need a great deal of other stats to adequately judge a players worth. You can't use BA and decide how much a player helped a team. Using HR and RBI doesn't give you a great picture of a player's contribution. It's great because it is essentially a rate stat (like OBP or SLG) but has been altered to work like a counting stat (like HR or runs). You don't have to try to guess the impact he made without knowing info like PA. You don't have to see a stat like 10 HR and not be impressed just because you didn't know it was over only 65 at bats. It allows you to know how much contribution a player should have added with his play over the amount of time he actually player. It might not be as good as a stat like WPA in determining actual contributions but it is much more predictive and avoids the problem of judging a player's worth based on good or bad luck.

As a fan, I obviously can't deal with just one stat. Like many or probably most on here, even before creators of these advanced stats were born I would dive as deep as I could into stats. That started as a child looking at the numbers on the back of my baseball card and being amazed at how easily I could compare the value of players that I didn't even really know. BA, RBI, and R are great but I definitely enjoyed when OBP and SLG came along and considered them welcome additions. I never stopped looking at a particular stat just because another one came along. Even now when looking to gauge a player I know that RBI and R are not a good way to determine worth. But I still look at them. I don't just skip past them. I've been looking at these numbers for so long it's impossible at this point to just simply ignore them. For me more numbers is always better.

But for a new fan of the sport just learning it and trying to understand the game I wonder if the more numbers the better mentality still holds true. I doubt it. More numbers would be confusing. They wouldn't let you really see the value a player has. For me I know RBI are worthless as a predictive stat but since I've been learning them for so long I can't just ignore them. But for a new fan, why would they want to use RBI to judge value? The stat isn't helpful and it seems like it would be mostly confusing to them to try to explain why we still look at it. Because there is no real answer. We just do. But with WAR they don't have to. They can look only at what provides value. It's definitely a little harder to calculate but to that I'd say two things: First, it's not impossible. The calculations are still arithmetic, not calculus. Second, you don't need to do the math anyway. How many sites show a players WAR now? Basically every stat site out there. Newspapers are carrying WAR now. Certainly some of the fun of stats as a kid was calculating them but that wasn't necessary for my enjoyment of the game.

So I say WAR isn't overused. Not more than any other stat. Many people ignore entire sections of stats if a player has a high OBP or hit a lot of home runs. They'll quote only the stats that help their point. This isn't new. But in my opinion, more so than any other stat, WAR is very close to equal with all of it's individual components. You could take all the single numbers that make up WAR and have only a slightly better picture of a player than you'd have by just WAR itself. That's something most other composite stats can't say.

If I could make some changes though, I'd say: A) Stop using pitching WAR. It's almost completely useless if it's from BR and if it's from Fangraphs it is completely useless. And B) I'd like people to breakdown oWAR and dWAR. dWAR at this point is still a little hard to trust and I'd like to know right off the bat if a player's value is tied mostly to their bat or their glove. If it's tied to their glove it tells me I should do a little more digging to get a more accurate picture of the player. The given WAR may still match my view but I know it's far more likely to do so if the value is tied to the bat versus the glove. I'd like to know how much I should trust the number and I feel separating those two out gives me a better idea of that.


This was a bit 'ranty' but pretty good insight!

The 'old stats' have their place but they are not meaningful in serious discussions of today's game. They are for fun, much in the same way as 'game winning RBI's' or those stats I used to make up when I was younger. Cabrera winning the triple crown was seriously cool, but it should have had no merit in the MVP discussion.

As for your changes, I actually like pitcher WAR from fangraphs, just not as a serious indicator of total value. It is only looking at things the pitcher has direct control over and giving him credit for those. Good pitchers can be more valuable, bad pitchers less valuable, but this serves as a baseline that ignores the value of defense. It is trying to not overstep its bounds.

One change the stat could make is to include error bars to accompany the defensive metrics. That said, I'm not sure if they are actually needed. An outfielder could be +10 runs on defense in CF in just 30 innings. We know he wont keep that pace up, but it doesn't mean it didn't happen. It's not all that different from hitting 10 homeruns in a week. We know there are sample size issues, but the stat is not trying to predict actual skill level, but rather to simply report performance.

People simply need to understand that defensive value is far more reliant on chance than offense. Batters are going to get 3-6 PA's a night, but the number of balls hit to a CF is much more volatile. Defenders in the right position on the right team are going to have outlier defensive years as reported by WAR, and this isn't pure chance. It's more along the lines of a batter hitting 100 homeruns in a season because he got 1200 PA's. The value is real, even if unrepeatable.
Top of the page Bottom of the page
troutroy
Posted 2013-02-01 8:21 PM (#1932440 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse




Posts: 31


Most WAR discussions comes down to discriminating between offense and defense. I thought of a theory that discriminates slightly against defense that makes defense 7/10th's as important as offense, if that was true, Miguel Cabrera and Mike Trout's WARs would be just about even. Also there's opportunity's per game. So let's say a player averages 4 PAs per game, and 2 defensive opportunities per game. You get less chances to 'f it up" defensively. And all you need is 1 hit or walk on offense per 4 PAs a game to hold an above average on-base-percentage or average. Defense has less room for error percentage wise, obviously.
So the argument "defense isn't as important as offense" doesn't carry much weight. I'm only covering two aspects of the large equation, but defense vs. offense seem to be the topic of most arguments.

Edited by troutroy 2013-02-01 8:28 PM
Top of the page Bottom of the page
troutroy
Posted 2013-02-01 8:35 PM (#1932444 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse




Posts: 31

Baseball 101 states that offense outweighs defense because on defense defense, it's you and 8 other guys that can potentially field the ball, and when you're hitting it's just you and the pitcher. In my opinion, when it comes to defense , when the ball is headed your way it's just you and the ball. I believe the two are in the same light for a reason, because they are equally important.
Top of the page Bottom of the page
troutroy
Posted 2013-02-01 8:40 PM (#1932446 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse




Posts: 31

Sorry if my first post was confusing :P
Top of the page Bottom of the page
AZMike
Posted 2013-02-01 8:41 PM (#1932447 - in reply to #1932002)
Subject: Re: ESPN: Let's corral the WAR horse



All-Star

Posts: 1695

Location: Tuscon
HaloFan85 - 2013-02-01 11:57 AM

Soto Speaks - 2013-02-01 8:33 AM

Don't know what's funnier, the comments or the avatars of the people making the comments.

What's even funnier is the actual number of comments.  Ever since they made you log into facebook to post comments, the numbers have gone from 5000+ to less than 100 in most articles.


There's also a LOT less internet toughguy-ism / general jackassery now that for the most part people's real identities are tied to their posts as opposed to throwaway spam accounts.


Top of the page Bottom of the page
troutroy
Posted 2013-02-01 8:44 PM (#1932450 - in reply to #1932447)
Subject: Re: ESPN: Let's corral the WAR horse




Posts: 31

AZMike - 2013-02-01 8:41 PM

HaloFan85 - 2013-02-01 11:57 AM

Soto Speaks - 2013-02-01 8:33 AM

Don't know what's funnier, the comments or the avatars of the people making the comments.

What's even funnier is the actual number of comments.  Ever since they made you log into facebook to post comments, the numbers have gone from 5000+ to less than 100 in most articles.


There's also a LOT less internet toughguy-ism / general jackassery now that for the most part people's real identities are tied to their posts as opposed to throwaway spam accounts.




Hence trolls
Top of the page Bottom of the page
AZMike
Posted 2013-02-01 8:55 PM (#1932452 - in reply to #1931919)
Subject: Re: ESPN: Let's corral the WAR horse



All-Star

Posts: 1695

Location: Tuscon
Angelsjunky - 2013-02-01 9:39 AM

I agree that WAR doesn't work so well for pitchers. It also doesn't work so well for comparing pitchers with hitters in terms of overall value. For instance, using wopphil's example of Jered Weaver who had a 3.0 fWAR in 2012 - let's look at what fWAR reveals. Remember, Jered Weaver had a 2.81 ERA in 188.2 IP.

- Weaver was less than half as valuable as Felix Hernandez (6.1) and Justin Verlander (6.8)
- # of pitchers with higher fWAR than Jered Weaver: 34
- A sampling of pitchers with higher fWAR than Weaver (2.81 ERA in 188.2 IP): Ian Kennedy (4.02 ERA), Jon Lester (4.82 ERA), Jeff Samardzija (3.82 ERA), etc
- # of position players with high fWAR than Weaver: 71
- A sampling of players with higher fWAR than Weaver: Omar Infante (3.2), Neil Walker (3.3), Ben Revere (3.4), Mike Moustakas (3.5, and a .242/.296/.412 line!), etc

Some questions that arise:
1) Who would you rather have in your rotation, Felix and a replacement level starter or two Jered Weavers? According to fWAR the two groups are of equal value.
2) Who was more valuable to their respective teams last year, Jered Weaver or Mike Moutakas (or any number of players)?
3) Was Jered Weaver really the 35th best pitcher in baseball last year?
4) Was Jered Weaver really the 106th best player overall in baseball last year?
5) When will the madness end?!

I like WAR, I use it a lot - but something is wrong here.


I prefer Win Probability Added over WAR as a descriptive stat about what happened last year. It passes the "smell test" a lot better compared to what I actually saw on the field. I'll use 2 position players and 2 pitchers to illustrate my point:
WAR:

Mike Trout = 10.0
Justin Verlander = 6.8 (#1 pitcher in MLB)
Jered Weaver = 3.0
Howie Kendrick = 2.8 (nearly as valuable as Weaver )

WPA:

Mike Trout = 5.32 (still #1 player in MLB)
Justin Verlander = 4.04 (still #1 pitcher in MLB, but not by ludicrous margins)
Jered Weaver = 3.39
Howie Kendrick = -2.25


I'd say 5.6 win difference between Weaver and HK makes a LOT more sense than a 0.2 win difference, maybe that's just me.

Edited by AZMike 2013-02-01 8:57 PM
Top of the page Bottom of the page
HaloMagic
Posted 2013-02-01 8:55 PM (#1932453 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 5418

I don't know that most people think defense is less important than offense. It's just that it's seemingly easier to add value through offense than through defense. A runs saved is equal to a run scored. It's just a lot easier for a player to contribute to a run than it is for him to save one and that is most likely due to the chances received. A player with average offense but poor defense will add more value than a player who is poor on offense but possesses average defense. At least as we understand it today. I don't foresee it changing, though. Less opportunities to save runs seems to be a huge factor in defense being less impactful.
Top of the page Bottom of the page
AZMike
Posted 2013-02-01 9:03 PM (#1932455 - in reply to #1932453)
Subject: Re: ESPN: Let's corral the WAR horse



All-Star

Posts: 1695

Location: Tuscon
HaloMagic - 2013-02-01 8:55 PM

I don't know that most people think defense is less important than offense. It's just that it's seemingly easier to add value through offense than through defense. A runs saved is equal to a run scored. It's just a lot easier for a player to contribute to a run than it is for him to save one and that is most likely due to the chances received. A player with average offense but poor defense will add more value than a player who is poor on offense but possesses average defense. At least as we understand it today. I don't foresee it changing, though. Less opportunities to save runs seems to be a huge factor in defense being less impactful.


The problem with quantifying defense is how do you divide up credit between all the players involved.

Take the example of a guy grounding into a double play. How is credit for saving one or more runs divided up among the pitcher / shortstop / 2B/ 1B who all touched the ball in that situation? The pitcher plays a large role in keeping the ball down/ trying to induce groundball contact. The SS makes a great diving stop then the 2B/1B each make good catches under pressure to catch a series of rushed throws.
Top of the page Bottom of the page
HaloMagic
Posted 2013-02-01 9:45 PM (#1932479 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 5418

There has to be a certain expectation of a play being made. Regular fielding percentage uses errors based on score keepers judgement. Advanced stats use extensive info and base it off of the average player. Either way, if a throw was at the glove and delivered on time then almost no credit should be given to a fielder for simply catching it. The credit for the out should go to the person who fielded it originally. How much credit, in the case of advanced stats, depends on the judged level of difficulty. If a fielder makes that play only 30% of the time then 70% credit should go to the fielder. If the fielder misses the play then he should lose 30% credit. When a massive amount of info is gathered like in the case of BIS a really good baseline can be created. One of the problems with this method is fielder position. If an outfielder is playing in and the ball goes over his head he might miss it whereas someone who was playing straight up might have got to the ball. Sometimes this is the fielders fault for over estimating his still and taking bad positioning. This should probably count against the fielder, as if he was playing straight up he would have been expected to get to it and the decision to play in was his own. But often, even usually, a player is playing in a certain spot because of the manager. Which means if he misses a ball it's not entirely on him. He wasn't allowed to be in the right position to get it. I predict that this issue can be worked out as they work to fine tune and more towards a system that judges plays based not only on end location of ball but also starting location of the fielder. A truly large database of prior plays would be needed for this, though. But each game that passes gets us closer to that point.
Top of the page Bottom of the page
Mile High Dreams
Posted 2013-02-01 9:51 PM (#1932482 - in reply to #1931877)
Subject: Re: ESPN: Let's corral the WAR horse



Hall of Fame

Posts: 7978

TL;DR = WAR sucks
Top of the page Bottom of the page
Cespedes
Posted 2013-02-01 10:23 PM (#1932494 - in reply to #1932452)
Subject: Re: ESPN: Let's corral the WAR horse


Arkansas Travelers

Posts: 637

AZMike - 2013-02-01 8:55 PM

Angelsjunky - 2013-02-01 9:39 AM

I agree that WAR doesn't work so well for pitchers. It also doesn't work so well for comparing pitchers with hitters in terms of overall value. For instance, using wopphil's example of Jered Weaver who had a 3.0 fWAR in 2012 - let's look at what fWAR reveals. Remember, Jered Weaver had a 2.81 ERA in 188.2 IP.

- Weaver was less than half as valuable as Felix Hernandez (6.1) and Justin Verlander (6.8)
- # of pitchers with higher fWAR than Jered Weaver: 34
- A sampling of pitchers with higher fWAR than Weaver (2.81 ERA in 188.2 IP): Ian Kennedy (4.02 ERA), Jon Lester (4.82 ERA), Jeff Samardzija (3.82 ERA), etc
- # of position players with high fWAR than Weaver: 71
- A sampling of players with higher fWAR than Weaver: Omar Infante (3.2), Neil Walker (3.3), Ben Revere (3.4), Mike Moustakas (3.5, and a .242/.296/.412 line!), etc

Some questions that arise:
1) Who would you rather have in your rotation, Felix and a replacement level starter or two Jered Weavers? According to fWAR the two groups are of equal value.
2) Who was more valuable to their respective teams last year, Jered Weaver or Mike Moutakas (or any number of players)?
3) Was Jered Weaver really the 35th best pitcher in baseball last year?
4) Was Jered Weaver really the 106th best player overall in baseball last year?
5) When will the madness end?!

I like WAR, I use it a lot - but something is wrong here.


I prefer Win Probability Added over WAR as a descriptive stat about what happened last year. It passes the "smell test" a lot better compared to what I actually saw on the field. I'll use 2 position players and 2 pitchers to illustrate my point:
WAR:

Mike Trout = 10.0
Justin Verlander = 6.8 (#1 pitcher in MLB)
Jered Weaver = 3.0
Howie Kendrick = 2.8 (nearly as valuable as Weaver )

WPA:

Mike Trout = 5.32 (still #1 player in MLB)
Justin Verlander = 4.04 (still #1 pitcher in MLB, but not by ludicrous margins)
Jered Weaver = 3.39
Howie Kendrick = -2.25


I'd say 5.6 win difference between Weaver and HK makes a LOT more sense than a 0.2 win difference, maybe that's just me.


This means Howie was very unclutch.
Top of the page Bottom of the page
Jump to page : 1
Now viewing page 1 [50 messages per page]
Jump to forum :
Search this forum
Printer friendly version
E-mail a link to this thread

Registered to: Angelswin.com
(Delete all cookies set by this site)