Article: Why Do fWAR and bWAR See Certain Blue Jays So Differently?

[Leo Morgenstern]

The beauty of Wins Above Replacement, or WAR, lies in the metric's simplicity. It's one number that aims to sum up all of a player's quantifiable contributions on the baseball field. You can use WAR to compare sluggers to speedsters to defensive wizards. You can use it to compare hitters and pitchers. You can use it to compare players across every era of Major League Baseball, from the pre-modern period of the nineteenth century to the present day. 

WAR isn't perfect. It's inherently subjective and imprecise, but when understood and applied correctly, it is an incredibly useful tool in the baseball fan's statistical toolbox. 

Yet, while WAR might be simple to use, it's definitely not simple to calculate. What's more, different sources have varying opinions about the best way to make those calculations. That means you can find a stat called WAR on multiple websites, but unlike batting average or ERA, a player's WAR won't be the same number everywhere you look.

To make things all the more confusing, far too many newspapers, blogs, TV broadcasts, and social media accounts fail to specify which version they're using when they cite WAR. You might be watching a game and hear the commentator say a certain player has 3.5 WAR this season, and the next morning, when you're reading about the same player in the paper, you'll see that number has all of a sudden dropped down to 2.4. I couldn't blame anyone who has had that experience for thinking the whole concept is arbitrary and useless.

The complexity of calculating WAR, the numerous ways in which it's calculated, and a lack of clarity about those various calculations can be enough to put plenty of fans off the metric entirely. That's a shame, because the whole purpose of WAR is to make the fan experience simpler and easier. 

So, I thought it might be helpful to write about some of the Blue Jays players with the widest gaps between their FanGraphs WAR (fWAR) and Baseball Reference WAR (bWAR). Let me be clear, however, that I'm not here to say which version is better. I'll lay out the differences as they pertain to these three players, and you can decide for yourself which of fWAR or bWAR is the metric you prefer to use. 

Alejandro Kirk

• fWAR: 3.8 (Second among catchers)
• bWAR: 2.1 (12th among catchers)

Kirk is the player who initially inspired this article. FanGraphs will tell you he’s been one of the best catchers in the sport and one of the best players on the Blue Jays. Baseball Reference paints a much more mediocre picture. What’s the explanation?

When it comes to catchers, the biggest difference between the versions of WAR from FanGraphs and Baseball Reference is that fWAR incorporates pitch framing and bWAR does not. Pitch framing is the ability to convert borderline pitches into called strikes, and some catchers are much better at doing this than others. Kirk is one of them. FanGraphs gives him a lot of credit for that – 8.6 runs of credit, to be exact. This is the number one reason why his fWAR is so much higher than his bWAR.

Framing aside, the defensive metrics that FanGraphs uses to calculate WAR are higher on Kirk’s glovework than the metrics that Baseball Reference uses. In particular, fWAR values his blocking and stolen base prevention skills more highly than bWAR. 

FanGraphs and Baseball Reference also calculate positional adjustments differently. In simplest terms, positional adjustments account for the fact that some defensive positions are harder to play than others. FanGraphs values a full season (1,458 innings) of play behind the plate at an additional 12.5 runs, while Baseball Reference would only credit a catcher with 9.7 additional runs worth of value for the same number of defensive innings.

Finally, on the offensive side of the ball, Kirk’s bWAR suffers from his propensity to ground into double plays, while fWAR no longer penalizes players for GIDP.

Eric Lauer

• fWAR: 1.5 (Fourth among Blue Jays pitchers)
• bWAR: 2.7 (First among Blue Jays pitchers)

FanGraphs WAR will tell you Lauer has had a really solid year for the Blue Jays. Baseball Reference WAR will tell you he’d be a dark horse Cy Young contender if he could pitch at this pace over a full season. 

In other words, there’s a good chance that fWAR more accurately describes what you think about Lauer’s true talent, but his bWAR better reflects the success we’ve watched him enjoy.

That’s because fWAR is based on the metric FIP, or Fielding Independent Pitching. So, fWAR doesn’t care how many runs a pitcher gives up. Instead, it’s concerned with the elements of the game that a pitcher can more directly control: strikeouts, walks, hit-by-pitches, and home runs, as well as the number of innings he pitches. In all of those areas, Lauer has been more good than great, hence his good-not-great fWAR.

In contrast, bWAR is based on a pitcher’s RA/9, a.k.a. runs allowed per nine innings. The idea is to compare the actual number of runs the pitcher in question has allowed to how many runs a theoretical "average pitcher" would have given up in the same situations. Lauer’s RA/9 is three-quarters of a run lower than his FIP, and it ranks him among the top 10 AL pitchers (min. 80 IP). It would be almost impossible to imagine an "average pitcher" allowing fewer runs to cross the plate. 

Lauer’s WAR gap is the more extreme example, but a wide difference between FIP and RA/9 also explains why a pitcher like Yariel Rodríguez has 1.0 bWAR and only 0.1 fWAR.

Myles Straw

• fWAR: 1.6 
• bWAR: 2.3

The batting and baserunning components of both fWAR and bWAR largely agree that Straw has been a slightly below-average offensive player; his bat is a weakness, but when he manages to reach base, he compensates with terrific speed. 

The fielding components of both fWAR and bWAR also agree that Straw is a fabulous outfield defender. However, evaluating the extent of his greatness is where they differ. 

DRS (Defensive Runs Saved) is the fielding component of bWAR. According to this metric, Straw has been the best outfielder in baseball this year, at least on a per-inning basis (min. 400). He has racked up 14 DRS in 592.1 innings, mostly in center field. Pete Crow-Armstrong leads the majors with 20 DRS, but he’s played almost twice as many innings as Straw.

FRV (Statcast’s Fielding Run Value) also loves Straw’s glove, but not quite as much as DRS, and it’s FRV that feeds into fWAR. Straw’s 8 FRV is still elite, especially considering he’s only played about half a season, but the six-run difference between his DRS and FRV is the primary explanation as to why his bWAR is about three-quarters of a run higher than his fWAR.

View full article

[Spanky__99]

I prefer bWAR - RA9 for pitching and fWAR for bats.