FanPost

Battle of the WARs for the Phillies: Fangraphs v. Baseball Reference (to the DEATH)

[Editorial note: Front paged from the fanposts because this is an extremely useful comparison of WAR stats, something we use quite heavily on this site.]


You turkeys have probably also noticed there are variances in the WAR calculations between the good folks at Fangraphs and at Baseball Reference.

Because I love the information at both sites, I wondered if there was a reason to value one set of information over the other, so I decided to do a very rudimentary comparison of the two WARs, using 2011 Phillies players.

Some very cursory findings after the jump...

First: a messy, clunky comparison:

 

PLAYER :          fWAR; bWAR; MedianWAR

 

VICTORINO :       4.1; 3.2 = 3.7

ROLLINS :           3.3; 2.8 = 3.1

UTLEY :              2.2; 1.8 = 2.0

RUIZ :                 1.7; 1.9 = 1.8

POLANCO :        2.0; 0.9 = 1.5

HOWARD :         0.8; 1.6 = 1.2

MAYBERRY :     0.9; 0.5 = 0.7

MARTINEZ :        0.3; 0.4 = 0.4

BROWN :           0.2; 0.4 = 0.3

SARDINHA :       0.2; 0.3 = 0.3

ORR :                 0.0; 0.1 = 0.1

BELOW REPLACEMENT LEVEL:

GLOAD :           -0.1; -0.1 = -0.1

SCHNEIDER :   -0.2; -0.2 = -0.2

FRANCISCO :    0.2; -0.7 = -0.5

IBANEZ :           -1.0; -0.4 = -0.7

VALDEZ :          -0.7; -1.1 = -0.9

 

A few observations:

++ I know this information isn't in this table, but FG seems to weigh a lot more heavily on defensive stats than BR, which, I think, is why there is reasonably large difference b/w the respective WARs of Howard (0.8) and Polanco (1.1) (and possibly Victorino; 0.9).

My guess is that Polly gets the benefit of the weighted defensive stats to inflate his FG WAR, as his offensive stats are below-average in both calculators.

Also-- FG has Howard at a minus defender (-3.0 UZR) while BR has him as slightly plus (0.2 TZR)

++ Regardless of 1.6 (BR) or 0.8 (FG) WAR, these numbers = HOWARD CONTRACT BALLSUCK.

++ Michael Martinez (0.4 WAR) is the aggregate 8th most valuable position player on the team, better than eight other players, including regulars Domonic Brown (0.3), Raul Ibanez (-0.7).  For real.

++ Michael Martinez's performance is eating Wilson Valdez's performance for breakfast, according to both WARs-- Mini-Mart is b/w 1 and 1.5 wins better than Valdez.

++ Is Sardinha really a half-win better than Schneider?

++ Mayberry rates as a plus defender in FG (0.7 UZR) and slightly minus in BR (-0.1 TZR)

++ Francisco rates as a minus defender in FG (-3.1) and BR (-0.7)

++ The result is that Mayberry (0.7) has outplayed Francisco (-0.3) by about 1 win.

++ There are FOUR consensus minus defenders on the team: Brown, Francisco, Ibanez, and Valdez.

++ Ibanez and Valdez have totally sucked according to both WAR calculators, but one has sucked more than the other depending on which calculator you look at.

 

As for the pitchers:       fWar; bWAR; MedianWAR

HALLADAY :                   5.0; 4.8 = 4.9

HAMELS :                      4.0; 3.9 = 4.0

LEE :                              3.8; 3.9 = 3.9

WORLEY :                     1.2 ; 2.0 = 1.6

BASTARDO :                 0.9; 1.7 = 1.3

MADSON :                     0.9; 1.3 = 1.1

OSWALT :                      0.9; 0.9 = 0.9

KENDRICK :                  0.2; 0.9 = 0.6

STUTES :                      0.0; 0.5 = 0.3

CONTRERAS :              0.2; 0.1 = 0.2

BLANTON :                  0.4; -0.2 = 0.1

ROMERO :                   0.0; 0.2 = 0.1

PEREZ :                       0.1; 0.0 = 0.1

MATHIESON :               0.0; 0.1 = 0.1

VALDEZ :                     0.0; 0.1 = 0.1

BELOW REPLACEMENT LEVEL:

ZAGURSKI :               -0.1; 0.0 = -0.1

CARPENTER :          -0.1; -0.2 = -0.2

HERNDON :              -0.4; -0.1 = -0.3

BAEZ :                      -0.4; -0.8 = -0.6

 

++ My big take-away from this comparison is that BR definitely weights their pitching calculation more on results, whereas FG seems to weight more on predictors.

I say this because Worley, Bastardo, Madson, Kendrick, and Stutes all have a higher BR WAR, and each of these pitchers have outpitched their predictors (each have lower xFIP and SIERA than their actual ERA).

Conversely, Joe Blanton has a much better FG WAR (0.4) than BR WAR (-0.2), whereas he has definitely underperformed his predictors with a 5.50 ERA and a 3.46 xFIP and 3.54 SIERA.

++ Valdez has a higher WAR as a pitcher than as an infielder.

++ There are three consensus below-replacement pitchers: Carpenter, Herndon, and Baez.  To be fair, however, Carpenter has only pitched 6.1 innings and has pretty decent predictors (3.65 xFIP and 3.29 SIERA).

++ Halladay, Hamels, and Lee are VERY GOOD.

 

To conclude-- Baseball Reference seems to weight WAR calculations more on offensive stats for position players (and less for defense), and on results (sustainable or not) for pitchers (and less on results-predictors).

Fangraphs seem to weight more on defensive stats than Baseball Reference, and to reward pitchers more for peripherals and predictors than for actual runs allowed.

For these reasons, I feel am most likely to agree more with Baseball Reference's WAR calculations for position players (as defensive statistics are more volatile and arguably more difficult to quantify) but agree more with Fangraphs' WAR evaluations for pitchers, as I believe results-predictors such as xFIP and SIERA to be a more discerning tool for evaluating both present-moment and future pitching performance.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join The Good Phight

You must be a member of The Good Phight to participate.

We have our own Community Guidelines at The Good Phight. You should read them.

Join The Good Phight

You must be a member of The Good Phight to participate.

We have our own Community Guidelines at The Good Phight. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker