Wins Above Average as a Measure of Peak and True Longevity
Everybody knows what WAR is at this point. The 2012 and 2013 AL MVP debate brought it tons of publicity, and that was almost 10 years ago. While few people are able to explain the intricacies of the construct, I imagine that most big baseball fans can explain the gist of it even if they’re not a fan. Despite its flaws — and all models in all the sciences have flaws — it can be quite useful. However, its much lesser known cousin Wins Above Average (WAA) has its uses, too.
I’ve mentioned the stat in passing a few times at my former home at Beyond the Box Score, but I’m hard pressed to think of ever seeing a writer mention it anywhere else. The one big exception is the Hall of Stats, a fun site that evaluates baseball players’ Hall of Fame worthiness based solely on their stats.
Of course, nobody is arguing that this is the way it should be done, but it’s a fun, useful exercise to see how a player stacks up to the Hall of Fame standard when stripped away of our flawed human biases and subjectivity. You can say the same thing about Jay Jaffe’s JAWS system to a certain extent, but the Hall of Stats uses a different methodology that I find interesting. You can read all about it here, but what I found most interesting is there use of WAA as a measure of peak.
The Hall of Stats was started by a fellow former writer at Beyond the Box Score, Adam Darowski, who not too long ago tweeted out this fascinating tidbit comparing Fernando Tatís Jr. to Omar Vizquel, the subject of a previous OTBB statistical deep dive.
That shocked me. I had to double check, and sure enough, it’s true!
Tatís career WAA: 4.9
Vizquel career WAA: 5.3
Ignoring the fact that a rounding error makes it a 0.4 difference, how does a player with 45.6 WAR and 2,968 games played have only a slightly higher WAA than one with 6.9 WAR and 143 games played? Because it’s a lot harder to accumulate WAA than WAR.
WAR measures how much better a players is than a theoretical end of the bench, Quad A type player. WAA measure how much better you are than an average player. A 0 WAR means you stink, but a 0 WAA means you’re a solid player! The difference between the two is roughly two wins. Someone who is a subpar player year after year will accumulate some WAR but will constantly get negative WAA seasons.
This is part of what happened with Vizquel. When you take a career 83 wRC+, with only a few out of 24 seasons were you were an average hitter or better, and combine that with a good defensive shortstop, you’re going to get a player that is good at accumulating WAR but bad at accumulating WAA.
This is what I like about WAA: it penalizes compilers and players who pad their career lengths with bad play. Ken Griffey Jr., Albert Pujols, Rabbit Maranville, and Jim Kaat are a few examples of players who fall into that category. Pete Rose might be the worst offender here. In his nine seasons after leaving the Reds, he accumulated only 1.9 WAR at the cost of -12.7 WAA! Of course, he would’ve finished with “only” 3,164 hits had he decided to retire at age 37.
I’m firmly of the belief that we should evaluate players’ longevity not simply by how long they played, but for how long they were actually good. WAA provides a solid measure of this, as you can see how many seasons a player had a positive WAA value.
To be fair, the big discrepancy between Vizquel’s WAR and WAA has less to do with his longevity and more to do with another use of WAA: measuring a player’s peak. This is the bigger reason for Vizquel’s huge difference between his WAR and WAA. If a player has a long career where he is mostly good to very good, but not great, WAA is going to hold that against you. Tatís has been outstanding over his 143 games played. Vizquel, on the other hand, had only one season in his 24-year career where he had a WAR above 4.0 and a WAA above 2.0.
There’s no good way to do a search for WAR-WAA, unfortunately. However, if you peruse the JAWS leaderboards for each position, you’ll find that large WAR-WAA values were from those who played a long time but were never truly great or could never sustain greatness, played longer than they should have, or some kind of mixture of the two.
The most well known measure of a player’s peak is WAR7, which adds up a player’s best seven seasons by WAR, and is part of the JAWS system. WAA might just be another way to do that, but its use in measuring longevity compared to WAR is intriguing. I’ll leave you with a fun fact: Out side of his cup of coffee in 1993, Chipper Jones played 18 seasons and never had a WAA below 0.9. The Hall of Famer had a long career where he never had a bad season! Impressive!
