Ask LNH: Doesn’t kenpom treat all games equally?

This question came in response to our article regarding Wisconsin’s adjusted offensive efficiency. The Badgers now have a positive points-per-100 possessions adjustment of 7.0, or 5.8%, given to their actual per game average offensive efficiency. Both of these adjustments are larger than those given to any other team in the nation.

The answer to the question is “no.” When thinking about kenpom’s ranking system, always keep in mind that it’s a predictive system. Giving more weigh to recent games when predicting the immediate future doesn’t seem odd at all.

Let’s do some quick math to see if treating the Kentucky equally to all others would cause a different answer. Coming into the Final Four, Wisconsin’s AdjOE was 127.5. They had played 38 games. 127.5 * 38 = 4,845. After the Kentucky game, Wisconsin’s AdjOE is 128.5. An AdjOE of 128.5 * 39 games = 5,011.5.

If all games were treated equally, you’d expect Wisconsin’s AdjOE against Kentucky to be 166.5 (5,011.5 – 4,845). The Badgers scored 123.3 pp100p on a neutral court against Kentucky and their 86.5 AdjDE.

Continue reading Ask LNH: Doesn’t kenpom treat all games equally? →