fWAR, rWAR, and WARP

Wins Above Replacement (WAR) is a concept rather than one individual statistic. Batting average, for example, is defined as hits per at bat, but WAR is defined as the number of wins a player adds to his team compared to a replacement level player. Given that the definition for WAR is somewhat subjective, different groups of people have arrived at different calculations.

This page provides a basic overview of those differences. Please keep in mind that we do not speak for the other sites referenced in this post and if you want to know exactly how they calculate their metrics, you should inquire with them.

WAR Abbreviations:

The first thing you need to know about the different WAR values is that we use different abbreviations to refer to them. The three main WAR sites are FanGraphs, Baseball-Reference, and Baseball Prospectus. There are other versions out there, and plenty of analysts, teams, and consultants have their own blend.

FanGraphs WAR = fWAR

Baseball-Reference WAR = rWAR (or bWAR)

Baseball Prospectus WAR = WARP

WAR Basics:

The general framework for each method is essentially the same. You’re looking at hitting value, base running value, fielding value, positional adjustments, replacement level, and a few other small corrections for position players. For pitching, you’re looking at a run prevention statistic scaled to innings pitched. The difference among the metrics is the inputs more than the framework, but we can’t speak for everything inside the formulas as the other sites, so there may be some other conceptual differences. Here’s Baseball-Reference’s page.

Position Players:

Each site uses a different hitting, running, and fielding metrics. FanGraphs uses wOBA as it’s baseline hitting stat, but the other two sites use metrics built on a similar linear weights framework. For base running, each site has it’s own version of base running runs, which we abbreviate to BsR. For fielding since 2002, we use Ultimate Zone Rating, while B-Ref uses Defensive Runs Saved and BPro uses Fielding Runs Above Average. Prior to that, we use Total Zone, which appears to be B-Ref’s choice as well. Baseball Prospectus is the only site that has incorporated catcher framing into their calculations.

FanGraphs and Baseball-Reference use the same replacement level calculation of 1,000 WAR per MLB season. Baseball Prospectus’ appears to be slightly different.

Pitchers:

While WAR for position players is different among the three sites, it is nothing compared to the differences for pitchers. Generally speaking, the three sites take a run prevention estimate (more on this shortly) and scale it to how much the pitcher has pitched that season. The big difference is that each site use a much different run prevention estimate. Each site adjusts for league and park, so I’ll exclude that from the comparison for clarity.

FanGraphs uses Fielding Independent Pitching (FIP) as our baseline. However, given that infield fly balls are very similar to strikeouts in terms of being automatic outs, we treat them as strikeouts in our WAR calculation even though they aren’t include in the general FIP calculation. In addition, we include a leverage component when calculating our reliever WAR. I don’t know if/how the other sites use leverage for relievers.

Baseball-Reference uses a pitcher’s runs allowed and then adjusts that value based on the overall quality of their defense that year.

Baseball Prospectus uses their custom Deserved Runs Average (DRA) metric as a baseline. DRA is a significantly more complicated model that attempts to control for factors like catcher framing, defense, etc within the run estimator itself.

Things To Know:

The fact that there are different WAR values is not a problem, but rather a feature. WAR attempts to summarize a player’s on-field value and that is a very complicated question. Given how complicated baseball is, you would expect that people would arrive at different solutions to the same problem.

Positional player WAR values typically only differ dramatically when the various systems disagree about a player’s defense. The hitting and running stats are different, but they usually aren’t different enough to significantly alter the values you see.

Pitchers, however, are valued very differently by the different systems. FIP is a linear weights based system that treats all balls in play as equally valuable and ignores sequencing. Baseball-Reference starts with runs allowed and works backwards. Baseball Prospectus uses a complex modeling system to attempt to derive the value of individual events while controlling for contextual factors. You have to decide which method is the one you prefer, although looking at each site is the best way to get a complete picture of the player.