WAR: What is it Good for?

By Chas

I’m a stats nerd. I admit it. Despite this, I’ve tried to control my desire to bring SABRmetric analysis into our discussions on The Pine. Smitty and I were throwing Win Shares around when advocating for candidates in our Mt. Otsego series a while back, but I don’t think anyone else here really embraced the statistic.

I did feel as though my introduction of adjusted ERA (ERA+) and OPS (OPS+) had some impact, and Casey even suggested that maybe we should do a brief series introducing our fellow pinesters to some of the advanced statistics that are being widely used in baseball analysis today.

So, here we are. I can’t resist any longer. I need to tell you about the statistic that has supplanted Win Shares as the most popular SABRmetric measure out there.

The popular statistic I’m referring to is Wins Above Replacement, or WAR, which attempts to determine how many wins a given player is worth to his team compared to a replacement player. In this case, a replacement player is defined by what’s commonly referred to as a AAAA player. That is, someone who has found success at the AAA level, but is still not good enough to be a major league regular. In other words, the type of player who would replace a major leaguer if he got injured.

For position players, WAR takes both offensive and defensive contribution into consideration. I’m not going to attempt to explain how it’s calculated. I’m just going to say that several other advanced metrics factor into it. If you are interested in learning a little more about how WAR is calculated, I suggest you start here. Or, if you’re feeling really ambitious, check out the FanGraphs Library, a great resource for all your SABRmetric (or sabermetric, whichever it is) curiosities.

What is important is that WAR represents a composite of one player’s value to his team in a single statistic, and expresses this in a way that we all can understand. What’s also important to your understanding of this metric is an idea of what’s considered an average, good, excellent rating, etc.

For a one-year value, two Wins Above Replacement is considered to be the equivalent of the average everyday player, five is an all-star season and eight is an MVP-type season. In fact, according to Adam Darowski, who’s doing some excellent work with WAR over on a site called Beyond the Box Score, the average MVP winner since the advent of the award has produced 7.62 WAR.

Speaking of Darowski, he’s developed his own spin on WAR, which he calls weighted WAR (or wWAR). He uses wWAR as a method for evaluating players’ careers and Hall of Fame worthiness, but we can save that discussion for another day.

Where WAR differs considerably from Bill James’s Win Shares system is that WAR values can be negative. This is an advantage WAR has over Win Shares, because it is possible for a player to actually hurt his team by playing, whereas with Win Shares the worst he can do is contribute zero value.

If we used WAR to determine last year’s Most Valuable Players, Evan Longoria and Albert Pujols would have won the awards. In fact, both of the actual winners ranked no better than 5th in their respective leagues in the statistic, with Shin-Soo Choo, Miguel Cabrera, Robinson Cano and Adrian Beltre also ranking higher than AL MVP Josh Hamilton; and Roy Halladay, Ubaldo Jimenez and Adrian Gonzalez  beating out NL MVP Joey Votto.

But, we know that statistics have limitations, and when it comes to an award like the MVP, they can’t always measure the impact of a player’s performance. However, WAR does tell us that the criminally underrated Choo (7.3) was much more valuable than Delmon Young, who ranked higher in the voting, but whose defensive deficiencies brought his WAR total down to 0.8.

WAR is certainly good for something, and that is providing a benchmark that helps us overcome the difficulties in comparing players with vastly different skill sets, where statistical comparisons don’t provide obvious conclusions.

It also can be converted to value in terms of contract dollars, simply by quantifying the value of a win. In that sense, WAR could actually be used to determine who represents baseball’s biggest bargain at his current salary–possibly Choo, who earned $461,100 last year–and who is…well, A.J. Burnett (2010 salary: $16.5 million; 2010 WAR: -0.1).