Covering the Front and Back Pages of the Newspaper
February 20, 2013
BASEBALL: Established Win Shares Levels, RIP
I have come to a decision, absent a really compelling reason to re-think it: after doing them for nine seasons, I'm retiring my annual division-by-division Established Win Shares Levels reports.
First, it's an enormous amount of work. For nine years, I've run a 23-man roster for each of MLB's 30 teams and an EWSL for each of those players: a total of 6,120 computations, each and every one of them done by hand-entering the annual Win Shares data in an Excel spreadsheet and applying an annually re-adjusted age adjustment. I don't have a database or an assistant; every single number is my own effort.
The EWSL reports are by far the most time-consuming thing I do on the blog all year. Most of my other research projects (in baseball or politics) are smaller in scale, and often - being historical studies - they're not on the same time pressured frame as rolling out all six divisional previews before the season's too far underway for them to be meaningful. And they have to be cranked out at the same time every year when I'm engaged in preparing my fantasy baseball drafts and doing my taxes (and in recent years, my dad's taxes and my brother's estate's taxes), plus it's usually a busy time of the year at work.
Second, they're behind the times. In 2004, we had fewer ways to use all-encompassing "Great Statistics" to evaluate each team in anything like a comprehensive manner. The first PECOTA projections were only unveiled in 2003. Win Shares were something of a new kid on the block, and even if EWSL was never high science, it was - I thought at the time - a bit innovative to run an established performance level with them to get a rough estimate of the established major league talent each team had on hand. At the time, it seemed like a way to add a new angle to the discussion and move it forward. And I do think that, if nothing else, the age-adjustment data I compiled over the years is something of use.
But the state of the art has advanced a lot since 2004. With a quick click of the mouse, you can gather far more sophisticated projections at Baseball Prospectus, Fangraphs, the Hardball Times, and elsewhere than any value that EWSL adds. Many of those projections are developed by teams of people with a lot more mathematical expertise, computing power and free time than I have at my disposal, and at some point it's time to stop playing John Henry against the machine and know when you can no longer keep up. Most everyone these days prefers the more precise Wins Above Replacement to Win Shares anyway, since Win Shares are really a measure of gross value, rather than marginal value compared to a replacement level player. EWSL just isn't worth the effort that goes into it, and while I'm grateful for the support of my readers over the years, I'm not sure it will really be missed all that much.
Third, and related to the prior point...I can get more bang for my time by blogging other things. It's no secret that in the past 4-5 years, while my place in the baseball blogosphere has receded, my profile ability to reach a large audience with my political writing (especially at RedState) has increased a lot. I have no intention of abandoning baseball writing - I'd like to do more pieces for Grantland, for example, and I have some ongoing statistical research projects - but realistically, whether it's sharpening other baseball pieces or writing about politics, I can put my blogging time to better use than the annual time vortex that is the EWSL reports.
Thanks again to everyone who's read, commented on or linked to my EWSL posts. But it's time to move on.
Posted by Baseball Crank at 2:50 PM | Baseball 2012-14 | Blog 2006-14 | Comments (3) | TrackBack (0)