Hall of Stats

Update: The Hall of Stats was accepted as a poster presentation!

I love this site. I really like to talk about it.

For that reason, I've officially sent a presentation abstract for SABR 43 in Philadelphia this summer. Below, you can read my abstract (I used 499 of my 500 words).

Title: The Hall of Stats: An alternate Hall of Fame populated by a mathematical formula.

The Hall of Stats was conceived because the Hall of Fame voting process has become a political nightmare. A massive backlog of worthy candidates is piling up—some because of association with PEDs (or simply suspicion), but some because voters just don’t realize how good they were. There seems to be a false perception of what the Hall of Fame actually is. It’s not all Ruth, Mathewson, Cobb, and Wagner. For every Walter Johnson in the Hall of Fame there’s a Jesse Haines. For every Henry Aaron there’s a Tommy McCarthy.

Should each player better than Haines and McCarthy get in? No. But a player shouldn’t have to be Babe Ruth—or even Bert Blyleven—to get into Cooperstown.

There are 208 players in the Hall of Fame based on their MLB careers. According to the Hall of Stats, Blyleven ranks 34th among eligible players. He should have breezed into the Hall, but instead it took fourteen tries. Curt Schilling ranks 43rd. Jeff Bagwell ranks 50th. Kenny Lofton—who received less than 5% of the vote—ranks 92nd. There’s no reason to keep these players out of the Hall of Fame… if you look at things objectively.

That’s what the Hall of Stats does. It ignores anything that happened off the field—PEDs, lifetime bans, MVP awards, etc. The Hall of Stats takes the current number of players in Cooperstown (208), kicks everybody out, and re-populates itself with the top 208 players according to a mathematical formula. What you get is an objective Hall free of politics, grandstanding, and double jeopardy.

The formula is based on Wins Above Replacement and Wins Above Average from Baseball-Reference. A series of adjustments are made to deal with shorter 19th century schedules, greater 19th century pitching workloads, the grueling act of catching, and more. The adjusted WAR component represents longevity while the adjusted WAA component represents peak. They are combined and indexed to 100 (and called Hall Rating) so the Hall of Stats borderline is represented by a Hall Rating of 100.

Babe Ruth has a Hall Rating of 404. Blyleven’s is 189. Lofton’s is a robust 132 while Hall of Famer McCarthy is merely 27. 70 of the 208 players in the Hall of Fame (just about one third) are removed from the Hall of Stats.

The Hall of Stats aims to show how run- and win-value statistics can be used to measure a player’s Hall of Fame case. It also visualizes what a “default Hall” would look like if it was populated simply by the numbers. Should numbers be the only arbiter of who gets into Cooperstown? Certainly not. The Hall of Stats, and its accompanying website, is merely meant to serve as a conversation starter. That objective starting point is one thing that’s sorely lacking in the Hall of Fame voting process today.

The Hall of Stats does all of this with an ultra-modern website that features every player in Major League history, run-value-based similarity scores, rankings for every player (overall and by position), and more.

I've given a number of talks before about web development, but never about baseball (except for an internal one inside my company). I really hope I get the chance to do it.