Using WAR

[absintheofmalaise]

The use of WAR as a stat to talk about player value is on the rise here. Because of this, we ask that posters who do use WAR to also let people know which version of WAR they are using. Baseball Reference and Fangraphs calculate WAR differently. Since WAR is a combination stat, it would also help tremendously if you’d break down both the offensive component and the defensive component whenever you post using WAR.

Fangraphs uses Park Adjusted Runs Above Average based on wOBA to calculate the offensive component and UZR to calculate the defensive component. You can learn more about it here and here.

Baseball Reference uses the Sean Smith’s WAR data. You can learn more about how it is calculated here and more about Sean Smith here.

Thanks

 

[absintheofmalaise]

One other thing about using WAR, since it's a combo stat like OPS, could you also break it down into its individual components like you would with BA/OBP/Slg/OPS. Using fWAR as an example. They use Batting/Base Running/Fielding/Replacement/Positional = RAR = WAR .
Pedroia's line today looks like this: 28.6/-0.8/17.4/22.5/2.2 = 69.9 = 7.4

 

[santadevil]

Just noticed this article by Jim Caple on ESPN.com, discussing WAR a little bit and how fangraphs and bbref calculate those values differently and why it may be confusing to fans.

 

[JohnnyK]

Just noticed this article by Jim Caple on ESPN.com, discussing WAR a little bit and how fangraphs and bbref calculate those values differently and why it may be confusing to fans.

And this now led to B-R and Fangraphs to unify their replacement level baselines. This also means that historical WAR figures have changed, and career WAR for many players have changed significantly.

 

[absintheofmalaise]

Wanted to post this here from the Passan thread in MLB.
 
Tango posted this on pitcher WAR today because of the Passan article and explains more of the differences between the two.
 

First, I’m thankful that Passan has talked [/size]about WAR, and written a good article about it.  I’m also thankful that he made some mistakes, because I’m sure there are plenty of other people who make those mistakes.  This gives me a chance to correct their misconceptions.  Here’s one that he said:
 
 
Passan - "Cueto in fWAR is mostly his home run rate – .84 per nine innings – and the apparent disregard of its calculation for innings pitched." ​
 
 
I’m not sure why he would think the IP is being disregarded, and why that would be apparent.  If you look at the list of all pitchers, you will quickly see that the more IP you have, the more WAR you have.  The BASIC formula for WAR is extremely simple:
WAR = (performance per unit of playing time MINUS baseline performance per unit of playing time) TIMES playing time

 

[absintheofmalaise]

Tango on positional adjustments.

 

[tims4wins]

I have a question on WAR. If you had a team WAR of 0.0, how many games would you win? In other words, is a "replacement level" team a 40 win team? a 60 win team?
 
Trying to understand how many total WAR you need to build a 90, 95, 100 win team.
 
Thanks.

 

[SumnerH]

I have a question on WAR. If you had a team WAR of 0.0, how many games would you win? In other words, is a "replacement level" team a 40 win team? a 60 win team?
 
Trying to understand how many total WAR you need to build a 90, 95, 100 win team.
 
Thanks.

http://www.fangraphs.com/library/misc/war/replacement-level/

Until early 2013, it used to be that FanGraphs and Baseball-Reference both calculated WAR using different values for replacement level. That has since been changed, and both sites now calculate replacement level the same way:

This new unified replacement level is now set at 1,000 WAR per 2,430 Major League games, which is the number of wins available in a 162 game season played by 30 teams. Or, an easier way to put it is that our new replacement level is now equal to a .294 winning percentage, which works out to 47.7 wins over a full season.

 

[tims4wins]

http://www.fangraphs.com/library/misc/war/replacement-level/
 

 
Follow up:
 
If a replacement player is the league minimum cost - $400K - then 25 of those guys cost a team $10M, for 47.7 wins.
 
If a team wants to win 98 games (call it 97.7), they need to add 50 wins.
 
How is the cost of a win then calculated? Seems to me that if it was $5M per win then they would need to add $250M in payroll to achieve 97.7 wins (or would it be $5M less the $400K salary of the replacement player = $4.6M per win, which still would cost $230M to get to 97.7 wins).
 
What am I missing?

 

[SumnerH]

Follow up:
 
If a replacement player is the league minimum cost - $400K - then 25 of those guys cost a team $10M, for 47.7 wins.
 
If a team wants to win 98 games (call it 97.7), they need to add 50 wins.
 
How is the cost of a win then calculated? Seems to me that if it was $5M per win then they would need to add $250M in payroll to achieve 97.7 wins (or would it be $5M less the $400K salary of the replacement player = $4.6M per win, which still would cost $230M to get to 97.7 wins).
 
What am I missing?

Pre-arb/pre-fa/cost-controlled players.

Also almost nobody has 97.7 projected wins; a 98 win team is more likely a lower-win talent team that got some luck.

 

[tims4wins]

Got it, so it's not so much that a win costs $5M, but an additional win on the open free agent market does?

 

[SumnerH]

Got it, so it's not so much that a win costs $5M, but an additional win on the open free agent market does?

 
That's right.  It's why cost-controlled/farm-grown players who are productive are so valuable.

 

[Snodgrass'Muff]

I think part of where you are getting hung up is that teams don't start with a base of 25 replacement level guys. Rosters are constructed with a combination of low cost pre-arb players, arb eligible players and free agents who have been signed and the best teams have few, if any replacement level players on their 25 man roster. Another way of looking at it is that very few playoff teams are going to be getting 0 WAR out of one of their roster spots in the first place. More likely, you are looking at a roster with a lot of average players (~2.0 WAR) and a healthy number of above average performers. Some will be paid like replacement level players, some are paid market rates, many are in between, and those costs balance out to something less than what you would calculate a 90 or 95 or whatever win team to cost based on the cost of a win.

 

[tims4wins]

Thanks to both of you.
 
Is there an article pointing out where the $5M per win calculation comes from (is it even $5M or am I making that up)?

 

[SumnerH]

Thanks to both of you.
 
Is there an article pointing out where the $5M per win calculation comes from (is it even $5M or am I making that up)?

It's empirical and varies by year.  I believe they add up the AAV of all contracts and the WAR they achieve to figure out the current market value of a WAR.
 
http://www.fangraphs.com/blogs/the-cost-of-a-win-in-the-2014-off-season/

 

[tims4wins]

Thanks again

 

[Reverend]

Thanks to both of you.
 
Is there an article pointing out where the $5M per win calculation comes from (is it even $5M or am I making that up)?

 
It was estimated at $5m a few years ago. Many people believe it is significantly higher now.