This is a preview of the Joe Sheehan Baseball Newsletter, an e-mail newsletter about all things baseball, featuring analysis and opinion about the game on and off the field from the perspective of the informed outsider.
You can subscribe to the newsletter for one year for $79.95
using your PayPal account or major credit card. (Zelle users, please email me for details.)
--
The Joe Sheehan Newsletter: Lesson Learned
Vol. 17, No. 110
December 31, 2025
The final Newsletter of the year is traditionally, if not consistently, a look back at what I learned over the prior 364 days. Often the greatest lessons of a year have little to do with baseball, especially for me over the last decade or so, and I’d say that’s the case for 2025. It was an odd, sometimes thrilling, sometimes frustrating trip around the sun, one that rivaled the year the Mets had for peaks and troughs.
As far as baseball goes, I find myself thinking about how I have lived through an entire era from start to finish, and am seeing the start of the next one.
When I started talking baseball online, back in the early 1990s with the people who would create Baseball Prospectus, the principles were fairly simple. We wanted to shift the focus from context-sensitive numbers to context-neutral ones. From runs and RBI and wins and saves to ERA and OBP and SLG and strikeout-to-walk ratio. That was one of the first battles, and it took a while, but we did win that one. It was never about “killing” the win or the RBI, just diminishing their meaning relative to better measures of performance. Today, that victory shows up when television chyrons or the ballpark scoreboard or even your local paper runs AVG/OBP/SLG to describe how a player hit, versus AVG/HR/RBI. It doesn’t seem like much, especially for you readers too young to remember when the latter was dominant.
It’s also a dead letter now.
I think about the work I do, the work I have done just this week, and the slash lines that for years were the best way to describe performance are barely part of what I do anymore. Oh, they work in shorthand, same as the old ones did. “Jeff McNeil hit .243/.335/.411” tells me more than “Jeff McNeil hit .243 with 12 homers and 54 RBI” does. There’s more information, more good data, in the first line than in the second. That’s still a win.
For a long time, I’d have used only the first line to describe what McNeil did and who he is as a hitter. OPS was an even quicker shorthand. Alex Bregman is an 800 OPS guy. Kyle Tucker has 1000 OPS upside. I can’t remember the last time I made more than a passing reference to OPS, which was in many ways the gateway stat for casual fans into more advanced analysis.
No, these numbers, which I worked very hard to bring to use, now hide more information than they reveal. Three decades on from those first fights, we have data that go so much deeper, that tell us about how a player arrives at those three figures, what his skills are, and whether he’s likely to improve upon or decline from his output stats.
This hit me when breaking down McNeil yesterday, actually. I just wasn’t interested in either his AVG/HR/RBI or his AVG/OBP/SLG. My focus was on the things he did to get to those numbers. His pull rate. His flyball rate. His exit velocities. Looking at Munetaka Murakami last week, I focused on how often he made contact with pitches in the strike zone. When I wrote up Michael King’s return to the Padres, I mentioned his great change-up, which generated a 29% whiff rate and a .181 batting average against. Go look at the analysis being done with the swing data MLB released in 2024, how fast and at what angles players swing the bat, or the 2025 data on where players situate themselves in the batter’s box, stuff that I’m still getting my head around.
The transition here is different from the first one. AVG/HR/RBI was never the best information about a hitter, and over time it was replaced by better data. AVG/OBP/SLG -- or its pitcher analogue, ERA/K%/K-BB% -- is still telling you what kind of output a player generates, and doing it in a way that gets you much closer to the player’s contribution to winning. This isn’t about good or bad, but good and better. The output statistics are no longer where the conversations about players are happening. Now those are about inputs -- skills, movements, grips, stances, approaches. We want to understand how the players achieve those outputs, which means knowing things we, frankly, couldn’t have dreamed of knowing 25 years ago.
I’ve done this long enough to see the creation and popularization of an entire new vocabulary, and to see that vocabulary go extinct and be replaced by a new one. The acceleration of change in a sport that had to be dragged into the 20th century is remarkable.
What I learned this year is that the anchors of my work these last 30 years have come loose. Oh, I’m sure I’ll still use them because they are descriptors, because they’re good shorthand, because you all know what .209/.287/.401 means, and that a 2.87 FIP with a 28% K-BB is a great season. When it comes to real analysis, though, that’s not enough. We now have better data, data that does a better job of describing what Jeff McNeil and Munetaka Murakami and Michael King do to create and prevent runs, to win games for their teams. That’s the data that has to fill these pages in 2026 and beyond.
Well, until even better data comes along to replace it.
