This article is free for anyone to read, but please consider becoming a Patreon subscriber to allow me to keep writing posts like this one. Sign up to receive articles like this one in your inbox here.
On Thursday, FanGraphs ran a piece explaining that they were making a significant change to how they calculated wins above replacement. There’s nothing wrong with doing so, of course — it’s good that FanGraphs is making a change to how they measure defensive value, especially since what they are changing has been a known issue for some time now amongst people who pay attention to such things. These kinds of changes are how we end up with better understanding of which players are the most valuable, the least valuable, and so on. WAR isn’t the be-all, end-all, even if some treat it as such, but it can still be useful for analysis, so prioritizing its accuracy is necessary.
It’s not a static figure, though: historical WAR changes when the inputs change. Which is why FanGraphs’ list includes a number of tables that do things like show that Nick Ahmed has actually been worth nearly seven more wins above replacement in the last five full seasons plus the pandemic-shortened 2020 campaign than they were giving him credit for: from 2016 through 2021, Ahmed was rated as producing 4.8 WAR, but the change to the formula now has him at 11.6 over the same time period. Baseball-Reference, for what it’s worth, already had Ahmed at 11.5 WAR from 2016 through 2021, largely on the strength of his defensive output.
Now, the Players Association and Major League Baseball supposedly aren’t using FanGraphs’ WAR to determine the bonuses for pre-arbitration performers that were negotiated into the new collective bargaining agreement. They also aren’t using Baseball-Reference’s WAR: the full text of the CBA isn’t available to peruse just yet, so it’s difficult to see just what it is that’s going into the WAR calculations. Given they aren’t just using preexisting WAR figures for this system, it’s likely they’re figuring out their own measurement: whether it’s figuring out an average of the WAR calculations that are already out there, utilizing something a team or teams or the league itself are already using internally, or something new entirely, the problems inherent in this WAR-based bonus system remain.
WAR, as said, is not static: it is ever-changing, as new revelations in measuring performance come to light, which means that we’re always needing to look at legacy players and performances in a new light, whether that light is flattering or not. Which in turn means that basing a pre-arbitration bonus system off of whatever version of WAR ends up being utilized is just going to create a significant mess that, as of the way things stand, won’t be cleaned up. Imagine if Nick Ahmed went through his pre-arb years with the bonus system in place, and the WAR variant used to measure his performance was more than a win off every year. He might miss out on receiving any bonus whatsoever, or receive hundreds of thousands of dollars less than he should have through the system that is supposed to reward him for providing significantly more value than his contract status allows him to be paid for producing.
And there is the other side of things, too: Didi Gregorius was rated as producing 14 wins above replacement from 2016-2021, but the changes to FanGraphs’ WAR now have him at 7.8 WAR over the same stretch. Gregorius, by virtue of bad inputs, would be paid a bonus for value he wasn’t actually providing — money that, at least philosophically speaking, belongs to another player. The one “positive” we have going here is that the PA and MLB are going to be doing all of this in secret, in the sense that it’s unlikely we get to see what the inputs are, that it all be part of a black box operation, etc. So we won’t know when an Ahmed or Gregorius situation occurs. Not knowing doesn’t mean it isn’t happening, though, which mostly just means we’re going to have a secret pile of unidentified seasons where players were paid a bonus they, in hindsight, weren’t supposed to get, or were not paid when they should have been.
I don’t have a solution. This is just the nature of WAR: it is not, despite how it is sometimes wielded, immutable, nor infallible. Maybe there should be some kind of smaller pool set aside to pay out retroactive bonuses for the Nick Ahmeds of the world, so that a messy system at least attempts to clean up its messes once it notices they’re there. “Just get rid of the WAR-based system” isn’t a realistic option, even if some of us would like it to be.
The Baseball Writers Association of America deciding they don’t like being volunteered to determine whether certain rookies receive an extra year of service time or not, though? That’s something that is an option. There were grumbles from individual writers and editors belonging to the organization at the time the PA and MLB were negotiating these anti-service-time-manipulation measures into the CBA, but we need more. We need a total stop to a system that hands writers this responsibility. The fact that voters have any say whatsoever in how long a player will be under team control is ridiculous, but it just kind of happened, anyway.
If a player and team want to negotiate in a clause for a bonus if the player receives MVP votes or whatever, then sure, cool, that’s just betting on yourself to earn the votes and the bonus. The system that MLB and the union devised here, though, is supposed to be a deterrent to service-time manipulation for players who are still under the initial uniform player contract, and it’s also a determining factor in who gets the largest bonuses from the pre-arbitration pool. It’s a terrible idea even without getting into the ethics of it, and I hope the BBWAA makes a huge stink about it the next time they have a membership-wide meeting where change is on the table.
Visit my Patreon to become a supporter and help me continue to write articles like this one.