vorosmccracken.com

Part 1 here.

One of the things I always wanted to do with DIPS was to come up with a bit more rigorous way of calculating runs via only DIPS stats (BFP, HR, BB, SO, HBP). I’d always BSed it in the past, and the way I’m going to do it now is only slightly less BSed. But it’s a step forward anyway.

Tango’s got a lot of good stuff on David Smyth’s Base Runs. Base runs I always liked because it was both functionally accurate and theoretically accurate. It doesn’t work as well for individual hitters since hitters don’t get bat 30 times in a row. But it works fabulously for team hitting and individual and team pitching.

The formula has four basic parts A, B, C and D which combine like (A*(B/(B+C)))+D to give you your runs estimate. A represents base runners (minus home runs), C represents outs, D represents home runs and B represents an advancement factor. As such runs will equal home runs plus base runners multiplied by their ability to advance home.

You can read the formulae there, but for DIPS purposes we kind of have to make our own. I’ll spare you the linear regression details, the idea was to use a regression to come up with reasonable approximations of the A, B and C factors without using actual hits allowed totals (we already have D which is just home runs). The grid of the coefficients can be found here.

Using this new way of doing things, I list the actual Runs per 9, Earned Runs Per 9, DIPS Base Runs per 9 and DIPS Base Earned Runs per 9 for the three pitchers. I also list their league averages and then finally provide an ERA+ type of factor at the end (the DIPS runs divided by league average multiplied by 100).

                             DIPS    DIPS  league  league
Pitcher     R/9IP     ERA   R/9IP     ERA   R/9IP     ERA   FACT
Schilling    3.62    3.44    3.48    3.17    4.66    4.25    134
Brown        3.75    3.28    3.73    3.40    4.72    4.30    127
Mussina      3.88    3.63    3.92    3.57    4.99    4.58    127

Schilling comes out slightly ahead, and this makes some sense. His strikeout rate is superior and that gives him normal advantages. The DIPS ERA also gives him a HBIP advantage that his actual numbers didn’t actually show. He’s combined that with a very good walk rate and that means a very fine pitching career. Interesting Brown and Mussina come out more or less even but also with very good careers.

The next part will look at the post season and will examine whether any of them, some of them or all of them belong in the Hall of Fame.