WAR is Nonsense

[Punk]

Make love, not war.

[GregAlso]

At its basic level WAR compares how many RUNS a player is personally responsible for adding or preventing. Pitching and defensive WAR is still developing as a true understanding of what is controllable or what a player is personally responsible for is not complete yet.

[GregAlso]

No Shane, not 24 different players but a player like those 24, for which those 24 players are examples of last year. Each year other players are examples of it. It is a player that can easily be had on the waiver wire but is valuable enough to be a major league backup/part time starter. Someone slightly better than a AAA player.

What other fixed point should we compare them to? Huh?

[GuliblGuy]

Obviously certain players can make a huge impact on a club...but can you turn that impact into an absolute quantifiable number?  I don't think so, not 100%.

[Tyler]

Lol I misread that last post. Hence the delete haha

But was that a response to my post?

[shellback]

If you agree that the team who out scores the other wins than any amount of runs a player adds through offense or takes away through defense helps determine the wins/losses of a team. If you don't agree then yes it's bogus.

The key word is helps. some of the new stats are great- like whip but this one is overused and its validity in my opinion is weak. just an opinion- if you like it then have at it!

[stormngt]

According to Bob Ryan.

 

Discuss.

I don't think War is nonsense, but I would agree if people would say it maybe over-rated.  There are definitely flaws with it.  One is of which Ryan discussed....this so called replacement player has no statistical measurement.

 

However, War is probably the only statitistic that takes into account both offense and defense when measuring a player.  And defense is just as important as offense.

 

As for Cabrera  being a more valuable player than Trout because "Tigers don't have many quality hitters", well I don't know about that.  The Angels offense was pretty anemic when he came to the the club.

[halodground]

If you aren't a fan of WAR what stat do you prefer to value players?

[Lifetime]

Obviously certain players can make a huge impact on a club...but can you turn that impact into an absolute quantifiable number? I don't think so, not 100%.

it's not meant to be a true quantifiable number or an absolute. It is a relative number using common replacement factors rather than specific real replacement values for a team.

[GregAlso]

I agree it's over relied upon (you can read my last blog and know that,) but its not useless. It is a point of reference to compare and start a conversation. Understanding it is helpful and gives overall perspective. No it isn't 100% perfect. Yes, a player helps & WAR helps us quantify that in terms of runs (10 runs=1 Win.) It is a start that helps frame a discussion, not end it. Understanding how it works helps know where it's weak just like understanding the weakness of traditional stats helps us too. There has yet to be developed a perfect stat. There are just a series of stats that we can look at to get a complete picture.

Shane, where you asking me in your post?

[Tyler]

No Shane, not 24 different players but a player like those 24, for which those 24 players are examples of last year. Each year other players are examples of it. It is a player that can easily be had on the waiver wire but is valuable enough to be a major league backup/part time starter. Someone slightly better than a AAA player.

What other fixed point should we compare them to? Huh?

Nobody, its a lost cause. Too many variables go into a players statistical output. There will never be a single wholesome baseball player evaluation measured by any kind of mathematics.

And we shouldn't accept WAR more-so than any other stat because it's the 'closest' thing to the best because that gap between where WAR is at and the ultimate stat that will never exist is at is infinite.

[GregAlso]

Shane, there has been some great advances through linear weights about the value of a players offense. That is definitely becoming possible to quantify and compare. It is defense and pitching that is difficult right now. That is where WAR can break down. As the variables are better understood the stat will greatly improve.

[gotbeer]

Imagine where the Angels would have been if we had Darvish instead of Weaver last year.  

[Tyler]

Until you guys figure it out, I'm gonna go play catch with a ball and a glove.

[GuliblGuy]

Until you guys figure it out, I'm gonna go play catch with a ball and a glove.

Please track your catch/miss #'s so we can calculate your WAR.

[shellback]

it's not meant to be a true quantifiable number or an absolute. It is a relative number using common replacement factors rather than specific real replacement values for a team.

hence wins over replacement. that clears everything up.

[Tyler]

No offense, no good.

[Lifetime]

hence wins over replacement. that clears everything up.

thanks for showing your ignorance.

[Corona]

I wouldn't say its nonsense but it has a lot of flaws, I mean how the hell did Matt Harrison have a 6.2 WAR while Weaver had a 3.7. That's just ridiculous.

[shellback]

thanks for showing your ignorance.

you're even cuter when you're angry

[Lifetime]

Yup I'm super angry, grrrr!

[Troll Daddy]

Imagine where the Angels would have been if we had Darvish instead of Weaver last year.  

 

In third place?

[gotbeer]

I wouldn't say its nonsense but it has a lot of flaws, I mean how the hell did Matt Harrison have a 6.2 WAR while Weaver had a 3.7. That's just ridiculous.

 

 

So if we had Harrison instead of Weaver, we would have been + 2.5 games.  Rangers would be -2.5 games.  Angels would have tied for the AL West title.  

[Corona]

So if we had Harrison instead of Weaver, we would have been + 2.5 games.  Rangers would be -2.5 games.  Angels would have tied for the AL West title.  

 

 

Texas should invest in some outfield rocks.

[Hubs]

My biggest issue with WAR is the fictional replacement player's stats. They are not defined anywhere, except to say that a team with all replacement level players would win around 52-53 games. 

 

That's not enough. I want to see what a 0 WAR player looks like in terms of average, on-base percentage, slugging, wOBA, etc. 

 

Secondly I don't like positional player adjustments. Where a player plays defensively has zero effect on his ability offensively…just in context to this fictional replacement player who plays all 8 positions. So why take away runs?

 

dWAR is subjective, true, which means I want to see offensive stats and defensive stats used more often in conjunction with WAR. I also don't understand why oWAR and dWAR don't add up to be a players WAR. It comes down to positional adjustment being used twice, but why is it used in offensive WAR? If my Second Baseman is my best hitter (Yankees) and my first baseman is as good offensively, why don't they have the similar oWAR? It penalizes players who provide offense at traditional offensive positions, and credits guys who have good years at traditionally bad offensive positions.

 

Here's the best example of what I am talking about. 

 

Ben Zobrist plays multiple positions for the Tampa Bay Rays. In 2011 he had a .269 average with a .353 OBP, a .469 slugging, 46 2B, 6 3B, 20 HR, 19 SB. He struck out 128 times and walked just 77 times. He played 131 times at 2nd, 38 in the OF and 3 at DH.

 

In 2012, Zobrist hit .270, with a .377 OBP, a .471 slugging, 39 2B, 7 3B, 20 HR, 14 2B. He struck out less, just 103 times and walked more at 97 times. The years are pretty comparable offensively, 2012 being a bit better than 2011 in most areas. The only lead for 2011 was SB and 2B.

 

Except as a 2B, his nearly identical stats are better than when he played 47 games at SS, 58 games at 2B, and 71 games in the OF in 2012? 

 

His defense wasn't as good, primarily because of his time at short, where he was still pretty good with 6 errors in 47 games…and he played more at a traditional power position in the OF (33 games more).

 

How then is a superior offensive season have a lower oWAR? Because of where he played. Which makes zero sense. His defensive WAR went way down from 3.2 in 2011 to 0.4 last year, when he was arguably more valuable for being able to play SS, 2B, and the OF. He didn't play any of the positions badly, but to say he was worth three less wins 8.5 in 2011 to 5.5 in 2012 is ridiculous. 5 less steals, 7 less doubles, and a few more errors do not make a 3 win difference in real life, but apparently they do. His fielding contributions as an OF and playing some short should not have made that big of a difference, but they do.

 

I'd rather see the positional adjustment removed from the formula, use offensive numbers as offensive numbers, then provide a fielding WAR that an be added to a team to add or subtract to get a players total WAR.