Wednesday, June 15, 2011

Basketball Effective FG Percentage, and Other Statistical Modernizations

Backwardness offends me, especially when there is absolutely no reason or justification for it. Happens often with popular "statistics". I'm not talking about mildly complicated statistics, as might be taught in an intro stats class. I am talking about very basic numbers and metrics.

Basketball Field-Goal Percentage
The 3-point field goal has been part of mainstream basketball for about 30 years. It is an important aspect of the game. Yet the ancient statistic of field-goal percentage--originating before the 3-pointer had been dreamt of--is still the common yardstick of shooting performance. Why?! The more useful statistic would be the "effective field goal percentage", which simply weights the 3-pointer 50% more. It is SO obvious, I do it mentally whenever they flash a players 2 and 3-point stats.

GPAs
This has got to be one of the silliest traditions. Typically, grade-point averages (GPAs) are calculated using a scale where A=4, B=3, C=2, D=1, with possible interpolated weights for + and - grades. With ranges for each letter: A=90-100, B=80-89, C=70-79. So let's see...students take tests, and get numerical grades. Those numerical grades are summed and averaged. That result is then placed in a rough range that translates to a letter. Then, those letters are re-translated to a GPA. Crazy stuff--lots of extra computation, all the while removing resolution! The thing that drives me crazy is that this system is universally accepted by very smart people!!!

Stock Quotes
This is a bright spot. For years, stocks were quited in binary fractions--down to sixteenths, or perhaps thirty-seconds. Ridiculous stuff, especially in the age of the digital computer. Yet it persisted. This may be a special case, because I think part of the reason it persisted was entrenched interests--brokers and dealers  benefited from the rounding. Nevertheless, at some point I think in the 1990s, the practice was eliminated. So, progress is possible!

3 comments:

  1. Anonymous21:22

    Here, here.

    _Rocketstar

    ReplyDelete
  2. Eric - you've touched on a subject near and dear to my heart. I've been following tempo free statistics in basketball for a number of years. There are a slew of stats that commentators use that are irrelevant. Check out www.kenpom.com. He is the expert on advanced (read real) statistics in basketball

    ReplyDelete
  3. Well, as usual--great minds...! ;) Just checked out some of the tempo free stats, very interesting...PPP I especially like, right after effective free throws.

    I guess old habits of mind is here, as so often, the enemy of progress and truth. ;)

    ReplyDelete