They tell me that the guys with the best WAR are the best players. They also tell me that a great pitcher will win for you. OK, I’ll give them both of those (sorta). But one thing I’ve noticed is that they’re certainly no predictor of a championship. It’s the nature of the game that this would be true. You simply can’t let your ace pitcher (the one with the best WAR) pitch every inning and you can’t let your best hitter (again the one with the best WAR) come up for every at bat. It’s particularly true that you can’t take the guys with the best ever pitching WAR and find a lot of World Series championships.
I’ve been particularly critical of pitching WAR (but not as much critical of offensive WAR) ever since I saw the numbers and read the ever-changing formulae. But let’s accept that it’s a good measure of pitching excellence. It still isn’t much of a predictor of how a team will do. I Went down the BBREF list of yearly WAR (which uses BBREF’s version of WAR) looking only for pitchers. I excluded all pitchers who showed up before the advent of the 20th Century. In other words I ignored the pre-American League championship games (1884-1891). I did this because there is great disagreement about how seriously they were taken by the teams and players and how much they were treated as mere exhibitions. I also ignored the Temple Cup Series. Then I looked to find the top 10 WAR seasons for a pitcher in the American League era (1901-present). Of course I ran into Walter Johnson who had three of the top five and four of the top 12. So I changed the way I went at it. I began looking for a new name until I found 10 different pitchers. That took me all the way to 52nd on the list. Of course many of the 52 (and ties) were pre-1901 pitchers (including the first seven) and some were hitters (Ruth four times, Barry Bonds twice, and Gehrig, Yastrzemski and Hornsby once each). Here’s the list I ended up with: Walter Johnson in 1913 (16.0 WAR), Johnson in 1912 (14.6), Dwight Gooden in 1985 (13.2), Johnson in 1914 (13.0), Grover Cleveland Alexander in 1920 (12.8), Cy Young in 1901 (12.6), Steve Carlton in 1972 (12.5), Roger Clemens in 1997 (12.2), Johnson in 1915 (12.1), Fergie Jenkins in 1971 (12.0), Hal Newhouser in 1945 (12.0), Bob Gibson in 1968 (11.9), Alexander in 1916, Pedro Martinez in 2000, and Smokey Joe Wood in 1912 (all at 11.7). So the individual pitchers are Johnson, Gooden, Alexander, Young, Carlton, Clemens, Jenkins, Newhouser, Gibson, Martinez, and Wood (a total of 11).
Let’s notice a couple of things about this list. First, Walter Johnson’s 1912-1915 is, by WAR, the greatest pitching performance by a single pitcher over a period of years in the last 115 years (and people still debate how good he was). Second, there are a couple of one shot wonders in the list, specifically Gooden and Wood. The remainder are quality pitchers having their peak year.
But for my purpose, the most interesting thing is that only two of the pitchers were with teams that won the World Series: Newhouser and Wood. Gibson got to the Series but the Cardinals lost in seven games (Gibson himself taking the loss in game seven). In 1901 there was no Series, but Young’s Boston team finished second.
This isn’t a knock on pitching WAR, but merely an acknowledgement that it can’t predict pennants. And one great pitcher isn’t a predictor either. It does help if the number two pitcher on your team has a pretty good year also.