Win Probability Added
Another way to measure contributions that yield some surprising results
Win Probability Added
If you’re a regular here at CSJ, you’ve figured out by now that I’ve spent the offseason delving into different ways of measuring a player’s contribution.
The latest effort takes me to Win Probability Added, which fangraphs.com defines this way:
WPA (win probability added): WPA is the difference in win expectancy (WE) between the start of the play and the end of the play. That difference is then credited/debited to the batter and the pitcher. Over the course of the season, each players' WPA for individual plays is added up to get his season total WPA.
One of my favorite quotes related to sports is “Analytics see every play, our memories only see highlights.”
As fans we tend to defend “our guy” and remember when he does something good and forget the bad things. It’s human nature.
But there is a net of all the plays throughout a season (or career, for that matter) and some surprising names added more Win Probability than I expected, while others “stalwarts” did not.
This not only makes me think differently about those players, but also makes me curious about the stat, what it purports to show and if it really accomplishes the goal.
The Arms
Let’s start with the pitchers and the numbers below are for 2024.
Surprises:
*We all know Ryan Pressly struggled last season, but to be 24th on this list shows just how much. Sure, part of it was his role, where theoretically your team is ahead late and you are just adding incremental win probability. But what about Seth Martinez, Hector Neris and Shawn Dubin?
*When Justin Verlander signed with the Giants I mentioned to my son that they overpaid, in my opinion. He was not good down the stretch and was hurt multiple times.
That said I could use this as exhibit A.
*Abreu always makes things interesting, but he was nails overall, trailing just Hader out of the bullpen.
*This is another stat that shows why the Kikuchi trade was the right thing to do. In 10 starts Yusei accumulated the 4th highest WPA among starters and the top WPA per start average at .106, just ahead of Framber.
*The Astros need big improvement from Arrighetti.
The Bats
The batters were even more interesting to look at.
*Tucker was clutch and they are going to miss him.
*The first surprise comes at No. 4, as almost every other metric shows Jon Singleton as a liability. Not only am I shocked, given that every other metric, along with the Astros brass, felt the need to spend $60,000,000 at first base tells me Singleton was universally seen as a drag on the team.
*Yainer had good overall stats and his various splits look good, so it’s a mystery as to why Caratini added more net Win Probability and Cesar Salazar netted just as much with only 32 plate appearances.
Of course, a few plate appearances can benefit you if you have a big hit up front.
Diaz hit .313 with runners on,.346 with RISP, .405 with men on third and less than two outs, .350 late and close and on and on. It’s a mystery to me and along with Singleton makes me question the measure.
*Alex Bregman at 20th is amazing and shows you just how much he struggled in certain situations in 2024. Bregman hit .237 with runners on, .237 with RISP, .194 with 2 outs RISP, and .220 late and close.
*Jake Meyers and Jeremy Pena below Jose Abreu speaks volumes and I’ve documented Pena’s defensive regression, but this is an indictment of him offensively.
For reference, Christian Walker was 0.14 and Isaac Paredes came in at 0.87.
Useful Stat or Noise?
I’m new to studying Win Probability Added and need to do some deep dive before I come to any conclusion, but I have questions.
The top of the 2024 list for batters looks solid: Ohtani, Judge, Soto, Profar, Witt, Jr.
But then we have .278 hitting Corey Seager ahead of Yordan, despite Yordan having higher numbers across the board, some that were not even close, including 0.8 more fWAR.
That could be an outlier, but the cases of Yainer and Singleton, one overly generous to the positive and one overly harsh to the negative have doubt creeping in.
That leaves me unsure about a stat that I thought would be pretty straightforward, but I’ll track it closer this season and form an opinion on whether it’s useful or not.
Thanks for reading!