Peaking at the Margins: Washington’s Luck and Other Pac-12 Points
Posted by Adam Butler on January 10th, 2018According to KenPom’s luck rating, Washington is the most Irish team in the country. Let’s now dissociate ourselves from preconceived notions of luck and consider both what the rating means (you can read about it here) and more relatable measures of success and expectations. To synopsize the luck statistic, this is a measure of a team’s success relative to expectations (as established by their efficiency ratings). At 12-4 overall, with an adjusted efficiency differential of just +5.44, the Huskies sit ahead of similarly efficient teams in terms of wins and losses (although most have between one to three more losses). Consequently, they rate 108th by KenPom, or slightly above average. What do other margins say about the rest of the Pac-12? As we’ve established, the Huskies are the luckiest team, but what else can 17 or so games tell us about the remaining and critical two-and-a-half months of Pac-12 Hoop?
To start the conversation, I examined scoring differential (total points for minus points against), the Synergy Sports points per possession margins and the aforementioned KenPom adjusted efficiency margins. The Pac-12 results:
Team | W | L | Scoring difference | PPP difference | KP difference |
Arizona | 12 | 4 | 171 | 0.127 | 20.17 |
Arizona State | 13 | 2 | 252 | 0.185 | 20.08 |
UCLA | 12 | 4 | 130 | 0.100 | 13.91 |
USC | 11 | 6 | 142 | 0.078 | 13.87 |
Utah | 10 | 5 | 121 | 0.122 | 11.88 |
Oregon | 11 | 5 | 179 | 0.125 | 10.66 |
Washington | 12 | 4 | 55 | 0.050 | 5.34 |
Colorado | 10 | 6 | 28 | 0.016 | 4.89 |
Oregon State | 10 | 5 | 95 | 0.077 | 4.6 |
Stanford | 8 | 8 | -6 | 0.002 | 4.34 |
Washington State | 8 | 7 | 1 | 0.043 | 0.51 |
California | 7 | 9 | -84 | -0.088 | -2.59 |
Some quick notes followed by a few takeaways: 1) scoring differential is a predictor of success and usually an indicator that you’re a really good team regardless of record, 2) The PPP difference column is based on Synergy data, which accounts raw points per possession information into its metric (i.e., it’s neither a prediction of possessions nor adjusted for strength of schedule, home/away or otherwise), 3) a reminder that the KenPom difference includes the aforementioned adjustments.