a question about WAR

Printable View

11-14-2012, 12:03 PM
benchwarmerjim

a question about WAR

I got a qustion about WAR. Baseball-Reference has Mike Trout at a WAR of 10.7 and Fangraphs and 10.0. Why are the numbers different?

There has been a lot of mockery of baseball writers for not accepting the new way of number crunching and voting Miggy because he won a Triple Crown and using 'traditional' stats. Well, how are the old guard supposed to come into the new way of thinking if no one can get the numbers right?

Im not against the new way of thinking about baseball (although there are so many stats out there, its a bits overwhelming).
11-14-2012, 12:08 PM
USAFChief

Fangraphs and BR pulled different formulas out of a hat when they invented WAR. They also use different invented defensive metrics.
11-14-2012, 12:44 PM
Brock Beauchamp

Quote:

Originally Posted by benchwarmerjim

I got a qustion about WAR. Baseball-Reference has Mike Trout at a WAR of 10.7 and Fangraphs and 10.0. Why are the numbers different?

There has been a lot of mockery of baseball writers for not accepting the new way of number crunching and voting Miggy because he won a Triple Crown and using 'traditional' stats. Well, how are the old guard supposed to come into the new way of thinking if no one can get the numbers right?

Im not against the new way of thinking about baseball (although there are so many stats out there, its a bits overwhelming).

The majority of the difference comes from the defensive side of the ball. BB-Ref uses DRS for their defensive ranking. Fangraphs uses UZR for theirs.

WAR has its issues and should not be used to exactly define a player's worth. A 3.5 WAR player is not necessarily worse than a 3.8 WAR player. The metric is still being refined and that's why we see differing opinions and stats from different organizations. WAR works best when used in large numbers (multiple seasons) and when used as a general guideline for a player's performance, not a hard-and-fast rule that perfectly defines their level of play. A 5 WAR player is certainly better than a 2 WAR player but that might not be defined by looking purely at batting statistics. But a 2.1 WAR player may not be significantly different enough from a 2.4 WAR player to matter. Take the metric with a grain of salt. Overall, I like the metric but feel that it still has a ways to go before it's entirely accurate, especially in small sample sizes.
11-14-2012, 12:55 PM
ThePuck

Quote:

Originally Posted by USAFChief

Fangraphs and BR pulled different formulas out of a hat when they invented WAR. They also use different invented defensive metrics.

Remember when everyone was convinced the world was flat...good times...
11-14-2012, 01:00 PM
snepp

Quote:

Originally Posted by ThePuck

Remember when everyone was convinced the world was flat...good times...

Don't mind Chief. He's perfectly open to, and objective on, 99.99% of subjects.

This is the .01.
11-14-2012, 01:02 PM
ThePuck

Quote:

Originally Posted by snepp

Quote:

Originally Posted by ThePuck

Remember when everyone was convinced the world was flat...good times...

Don't mind Chief. He's perfectly open to, and objective on, 99.99% of subjects.

This is the .01.

I thought I was just being funny :-) Having said that, I don't think they pulled formula's out their behind and just made stuff up either... No one wants to accept it now. We'll see how people think 10, 15 years down the road.
11-14-2012, 01:03 PM
snepp

Quote:

Originally Posted by ThePuck

I thought I was just being funny :-)

WAR is serious business, no time for funny.
11-14-2012, 01:05 PM
benchwarmerjim

Quote:

Originally Posted by Brock Beauchamp

The majority of the difference comes from the defensive side of the ball. BB-Ref uses DRS for their defensive ranking. Fangraphs uses UZR for theirs.

ahh.
shouldnt there be a ruling by some third party arbitrator to decide what the standard formula should be?
11-14-2012, 01:05 PM
ThePuck

Quote:

Originally Posted by snepp

Quote:

Originally Posted by ThePuck

I thought I was just being funny :-)

WAR is serious business, no time for funny.

If he truly is a former Chief, he'll appreciate a joking jab from another former AF SNCO...
11-14-2012, 01:07 PM
ThePuck

Quote:

Originally Posted by benchwarmerjim

Quote:

Originally Posted by Brock Beauchamp

The majority of the difference comes from the defensive side of the ball. BB-Ref uses DRS for their defensive ranking. Fangraphs uses UZR for theirs.

ahh.
shouldnt there be a ruling by some third party arbitrator to decide what the standard formula should be?

I always prefer fangraphs myself...they seem to be just a wee bit more accurate, IMO.
11-14-2012, 01:31 PM
Brad Swanson

FanGraphs has a pretty extensive look at the how and why of their version of WAR: http://www.fangraphs.com/library/index.php/misc/war/
11-14-2012, 02:18 PM
Brock Beauchamp

Quote:

Originally Posted by benchwarmerjim

Quote:

Originally Posted by Brock Beauchamp

The majority of the difference comes from the defensive side of the ball. BB-Ref uses DRS for their defensive ranking. Fangraphs uses UZR for theirs.

ahh.
shouldnt there be a ruling by some third party arbitrator to decide what the standard formula should be?

Early in the process, having competing metrics is a good thing. In time, one will prove more accurate and the industry will move in that direction. The more people you have striving for the same goal with different approaches, the more likely you will be to uncover something that works.
11-14-2012, 05:51 PM
USAFChief

Quote:

Originally Posted by ThePuck

Quote:

Originally Posted by snepp

Quote:

Originally Posted by ThePuck

I thought I was just being funny :-)

WAR is serious business, no time for funny.

If he truly is a former Chief, he'll appreciate a joking jab from another former AF SNCO...

As a former USAF SNCO, surely you are aware of the two rules regarding Chiefs:

1. The Chief is never wrong.

2. In the impossibility the Chief is wrong, see rule 1.

As for you, Snepp...let's just say I think I'm the .01 that IS being objective on this subject.
11-14-2012, 08:27 PM
biggentleben

Quote:

Originally Posted by Brock Beauchamp

Quote:

Originally Posted by benchwarmerjim

Quote:

Originally Posted by Brock Beauchamp

The majority of the difference comes from the defensive side of the ball. BB-Ref uses DRS for their defensive ranking. Fangraphs uses UZR for theirs.

ahh.
shouldnt there be a ruling by some third party arbitrator to decide what the standard formula should be?

Early in the process, having competing metrics is a good thing. In time, one will prove more accurate and the industry will move in that direction. The more people you have striving for the same goal with different approaches, the more likely you will be to uncover something that works.

See Blu-Ray vs. HDDVD or Beta vs. VHS