Is WAR a Criminal Statistic?

I, Andrew, am here to report a crime. A serial crime. A WAR crime. The victim in the case is recognition. The suspect??? I will let you be the judge. Let’s begin with the foundation of this case: wins above replacement, otherwise known as "WAR ." WAR is defined by mlb.com as “a player's value in all facets of the game by deciphering how many more wins he's worth than a replacement-level player at his same position.” What makes up the formula for WAR, a stat developed by Sean Smith with BaseballProjection.com. ? According to mlb.com, it can vary. Let’s use what they have listed as our baseline.

Batters:

• (The number of runs above average a player is worth in his batting, baserunning, and fielding + adjustment for position + adjustment for league + the number of runs provided by a replacement-level player) / runs per win.

Pitchers:

• Different WAR computations use either RA/9 (runs allowed per nine innings) or FIP (fielding independent pitching).Those numbers are adjusted for league and ballpark. Then, using league averages, it is determined how many wins a pitcher was worth based on those numbers and his innings pitched total.

The goal of this metric, according to mlb.com, is to “quantify each player's value in terms of a specific numbers of wins. And because WAR factors in a positional adjustment, it is well suited for comparing players who man different defensive positions.”

As to the values themselves (baseballreference.com):

8.0+ = MVP quality

5.0 - 7.9 = All Star quality

2.0 – 4.9 = Starter

0 – 1.9 = Replacement level player.

After reading the definition and ratings, it got me thinking. If the WAR metric is supposed to point to the most valuable player at each position, then why does the Baseball Writers Association of America ("BBWAA") seem to disregard the data in MVP voting? It leads to the question, can we trust the WAR rating? To find the answers to my questions, and likely create even more, I went to baseballreference.com. Take for example, last year’s MVP voting. The AL was a two horse race, Judge versus Ohtani. The NL race followed a similar pattern with Goldschmidt leading the charge. Followed closely by Arenado and Machado. Judge sent his AL record 62nd ball over the fence into the Toronto bullpen to capture the win. Goldy just kept on hitting and his competitors could not keep up. As for the rest of the AL and NL final results, they tell an interesting story.

The beauty of baseballreference.com is you can manipulate the charts. Go ahead, try it out: (https://www.baseball-reference.com/awards/awards_2022.shtml#AL_MVP_voting_link) The standard sort is the BBWAA writers’ opinions on the players’ performances in terms of vote points. You would think that the WAR column should follow right along with their opinions, right? WRONG!

The BBWAA writers and the WAR metric both accurately picked the two best players in the American League. They were Aaron Judge (10.6) and Shohei Ohtani (9.6). After that, it gets wild. The next 4 MVP picks by the writers were as followed: YordanAlvarez (6.8), Jose Ramirez (6.0), Jose Altuve (5.1), and Andres Gimenez (7.4). Our WAR metric chose in order: Andres Gimenez, Yordan Alvarez, Dylan Cease (6.4), and Mike Trout(6.3).

As for the National League, the top 6 vote getters were as followed: Paul Goldschmidt (7.8), Manny Machado (6.8), Nolan Arenado (7.9), Freddie Freeman (5.9), Mookie Betts (6.4), and Austin Riley (6.5). If measured by WAR: Nolan Arenado, Paul Goldschmidt, Sandy Alcantara (7.8), Manny Machado, Austin Riley, and J.T. Realmuto (6.5).

You, as did I, might be wondering, why is this different? Cease and Alcantara magically appear out of thin air. Altuve and Machado received the most votes at their given positions, even though they were not the most valuable at their positions. The second-best player on the Indians received more votes than their best player. What is going on here?

There is a bias towards position players on good teams. Gimenez shows the biggest glitch in the matrix for the American League. He was more valuable than Altuve at 2B, providing 2.3 more wins and providing an additional 1.4 wins more than Ramirez, his own teammate. BUT finished behind both in MVP voting. Over in the NL, Arenado fell victim to the same crime. He was proven, by the WAR metric, to be the most valuable player at his position and on the Cardinals. Did that matter? NOPE!

As for the 2nd oddity, why did Cease and Alcantara not finish in the top 6 in BBWAA votes? They posted tremendous seasons and dominated 2022. But, Alcantara finished 10th in the NL and Cease finished tied for 22nd, otherwise known as dead last. Why? Well, there is a common belief in the baseball world that a pitcher should not be able to win MVP. How can they be that valuable if they only contribute towards roughly 19% of games during the regular season? Looking back 30 years, only 2 pitchers have won the award. The most recent being Clayton Kershaw in 2014. I understand the common argument but do not agree with it. A dominant pitcher has more control and influence in how the game will ultimately finish. A great example of this is Alcantara and Cease. “Sox/Marlins losing streak ends as Alcantara/Cease dominate. Final score 3-1/2-0.” To further make my point, look at the Angels when someone other than Ohtani is pitching. The recap/story always reads something like, “Trout continues to dominate, Ohtani rewrites history, and the Halo’s lose 9-5.”

What can we take from this? Is there value in the WAR metric? Or should we just stick with what history tells us to believe?

Disclaimer: I would like to state that the purpose of this article is not to discredit Mr. Smith or the writers within the BBWAA.

--Andrew Soliday