We’re going to be working to add functionality over the next few days (the most notable missing thing right now is our win expectancy based stats–they’re getting a bit more of an overhaul, but they should be back soon). And as with any computer software, there is always the potential for bugs. There is a saying that a lot of eyes make all bugs shallow, and there’s certainly a lot of you–as we hear from you about potential problems, we’ll tweak things as necessary to provide the best possible analysis to you.
As I’ve mentioned previously, we’ve been overhauling the entire back end that generates our statistics reports. For those of you that care–we’ve moved away from a tangled mass of Oracle SQL, FORTRAN code and a few other things into a MySQL database, with a little bit of scripting on the side.
So what does that mean for you (other than, of course, our tardiness in delivering the stats–for which we are deeply sorry)? Well, there are two things. One is that our stats do a much better job talking to each other now–this is how we’re able to have BRR featured in WARP, for instance. Now all of our park adjusted metrics use the same set of park adjustments, our replacement level metrics all use the same replacement level, etc. We’ve eliminated some redundancies–having Support Neutral Win-Loss and Expected Win-Loss at the same time was redundant and confusing, and the sort of thing we’ve strived greatly to avoid.
In fact, let’s use the new slimmed down set of support neutral stats as an example. We’ve eliminated the support-neutral value added tree of stats, which cuts down on our impressive alphabet soup problem. (The reason for this is that having a pitcher metric measuring wins above replacement player is why we have, uh, Wins Above Replacement Player–especially as we standardized the foundational elements like replacement level and the like, the two flavors of stats became even more superfluous.) We’ve standardized on one set of advanced win-loss records, Support Neutral Wins and Losses. And we have those old standbys, Quality Starts and Blown Quality Starts. We’ve also added a new, BP flavored twist to the idea with Fair Quality Starts, derived from Fair RA. Unlike traditional Quality Starts, these adjust for a pitcher’s level of defensive support. They also use a baseline that changes along with the seasonal run environment and the park where the game is played.
The other thing it does is it provides a foundation for us to build on–now that we have the initial labor of building the new platform out of the way, the time to deliver new content on that platform is dramatically reduced from what it was in the past. So we’ll be rolling out some new functionality as the season progresses, and I think you’ll be excited once we get to that point.
On a last note–I'd like to thank everyone at BP, especially Dave Pease, Rob McQuown and Kathy Woolner, for their hard work in getting this done. And I'd like to thank you for your continued support of Baseball Prospectus.