Got an email from Ken Roberts of Sports Club Stats, a site that purports to determine playoff odds based on an algorithm Ken wrote. Being a bit of a stat-head, I was intrigued: cold, rational numbers are the security blanket I use to quantify emotional things in life like the results of baseball games. In fact, the reason I love baseball so much may be due in part to its close association with statistics as a meaningful measure of value. It's a control thing; some people have hordes of underlings or big houses to remind them who's in charge, I have baseball statistics. No wonder I run a blog.Anyway, I went to check it out. Ken's got a page for just about every professional sports permutation possible, which is cool - I could see how this could become a destination site for sports nuts if it proves to be reasonably accurate. The Red Sox page breaks things down into three big sections for the Home Towne Team: Today's Big Games (which games have the biggest effects on the Sox playoff hopes), Big Games (how games across the AL affected playoff chances for Boston) and What If (how many games the algorithm thinks Boston needs to win for a guaranteed playoff berth). Each section has explanations, but the page is complex enough to require quite a bit of exploration to start to understand.
One thing that jumped out at me: in the What If section, Ken records the team's chances of entering the playoffs based on their win/loss record over the remainder of the season and how many times that particular win/loss record came up over the millions of repetitions. The results play out in a bell curve, centered around a .500 record, which isn't surprising - .500 is supposed to be the average, after all - but that .500 record does predict a playoff berth no matter what, welcome news for all of us nervous fans. Next question: will Ken's method, which does not take match-ups, injuries or scores above 1 - 0 into account be at all accurate? I'll certainly be keeping my eye on his results.
Posts
