November 06, 2014

Bill Simmons being angry about a Mike and Mike segment talking about him out of context is a jump the shark moment. It's an intra-ESPN slapfight and nobody outside of the Beast of Bristol would care.

posted by rcade at 03:45 PM on November 06, 2014

Last night the Boston Celtics lost a game in which they committed 28 turnovers, yet they lost by only 3, and were in position to tie the game with a last-second three. The game could have been a Celtics blowout; rebounds, points-in-the-paint, shooting percentage, assists, all favored the Celtics. It led me to think about what sort of metric could be attached to turnovers, particularly turnover differential.

Last night's differential was around 18. My thought about how much it cost Boston is this: Multiply the turnover differential by the shooting percentage of the team with the negative differential (in this case Boston), round off, and add the result to the team's score. Do the same for the team with the positive differential, round off, and subtract the result from that team's score. This should give an approximate idea of what the score would have been with an even number of turnovers. Of course, turnovers will never come out even. Some adjustment should be made for steals, the assumption is that all of the scores that result will be 2 pointers, and there is no accounting made for free throws, but over all, this might be a useful metric.

grum will probably come up with something much more accurate, or probably tell me that I have re-invented the wheel but made it elliptical, but I'm still curious as how to apply some metric to the turnovers.

posted by Howard_T at 10:20 PM on November 06, 2014