clock menu more-arrow no yes

Filed under:

Does Northwestern deserve its KenPom rating?

New, 6 comments

Northwestern is 31st in KenPom. Fair or unfair?

NCAA Basketball: Maryland at Northwestern David Banks-USA TODAY Sports

It’s fair to say that Ken Pomeroy’s ranking system has been rather high on Northwestern all season.

Even when Northwestern was in the doldrums after losses to Purdue and Illinois, Pomeroy’s number-crunching never dropped the Wildcats below 32nd in his efficiency-based rankings. The Wildcats have impressively risen from 61st at the start of the year to 31st as of Thursday night.

But despite Northwestern’s big win over Wisconsin, the team has lost three of its last four games without Scottie Lindsey, and because Pomeroy doesn’t factor injuries or team morale into the rankings, it’s worth wondering whether No. 31 is a bit high.

The KenPom rankings have gone from the nerdy fringe to one of the most-cited rankings in college basketball over the past decade, and since the NCAA has dragged in “the analytics crowd” (Pomeroy, Jeff Sagarin, etc.) to give input on the NCAA Tournament Selection Committee for 2017, it’s as important as ever.

Debating whether Northwestern “deserves” to be 31st is a question of the mathematical formula for “adjusted efficiency margin” that Pomeroy uses to design his site.

Adjusted efficiency margin just takes a team’s offensive and defensive efficiency and compares it to the national average. Thus, rankings are solely based on how efficient the team is—no wins or losses are accounted for. Strength of schedule and home-court advantage are factored in. You can read more about the details here.

The results of these calculations lead to situations like the University of Virginia being ranked third in the KenPom rankings despite a 18-7 record. Almost no college basketball fans would argue the Virginia is the third-best team in the country, but because UVA remains one of the most efficient two-way teams in the nation, it is third.

The case of UVA could theoretically be applied to Northwestern as well. You see, according to Pomeroy himself in a Reddit comment, slow teams that play good defense are slightly overrated by the system. I did a simple calculation between adjusted efficiency ranking and pace for 2016-17 and found that there is indeed a slight correlation to playing slower and a higher KenPom rank, although you need a better sample size and fewer outliers to come up with anything definitive.

There is a slight correlation between having slower pace and ranking so far in 2016-17. Pomeroy said there was a causation, and thus there is also correlation.

Northwestern is a slow team that plays good defense. Anyone who’s watched any Northwestern game this year knows that. Northwestern is 26th in defensive efficiency and 283rd in tempo. Wisconsin, Cincinnati, Saint Mary’s and SMU all follow this same pattern and are perhaps slightly overrated (UVA, SMU and Wisconsin have been noted KenPom darlings c. 2015-present). That would partially explain how Northwestern has remained “sticky” in the high 30s despite its poor play earlier in the month.

However, I don’t think that Northwestern is overrated like UVA. For one, this measure is intended to be entirely predictive, rather than a qualitative analysis of the past. I think that Northwestern should still be thought of as around the 30th best team in the country going forward, especially with Lindsey. That makes sense.

Of the “four factors” (shooting, rebounding, turnovers, free throws) defensively, Northwestern has been incredible at limiting opposing shooting percentages. Northwestern is 13th in the nation in eFG% against, and that alone is enough to make up for its rather mediocre performance in forcing turnovers, limiting offensive boards, and conceding free throws (sidenote: Doesn’t it seem like the opposing team is in the double-bonus every half of every game?). Because shooting is by far the most important factor, Northwestern has built an elite defense through Pardon’s interior presence and excellent perimeter defense from Law, Lindsey and McIntosh.

I would contend Northwestern is 31st more for its defensive brilliance and decent strength of schedule rather than its offense. The Wildcats actually don’t play the analytical basketball as you might think given how dependent on three-pointers the team appears to be. Northwestern is 163rd in three-pointers attempted per game and are just above average in percentage of points scored from beyond the arc.

At this point in the game of college basketball, you simply must make threes to win games. Northwestern is no different. The team’s three-point percentage isn’t even that great compared to national averages now, with the team 184th in the country in three-point percentage (34.9 percent). That may have been good 15-20 years ago, but times have changed.

Of the four factors on offense, Northwestern only excels at avoiding turnovers. They are the 12th-best team in the country when it comes to limiting turnovers, which has kept the offense running even when the shooting isn’t there. As for the other two, Northwestern shoots at a high percentage from the line, but they hardly ever make it there. They are above average at offensive rebounding (99th in the nation), but not necessarily elite. The team’s offensive efficiency is now 55th in the nation, which I think is fair.

The combination of a reasonably tough schedule (KenPom rates 15-11 Wake Forest and 20-5 Dayton very highly), a slow tempo/good defense bump and good offense has Northwestern in 31st, somehow above Maryland (32) and just below Michigan (28). Northwestern’s close losses haven’t hurt it either, as the adjusted efficiency does not care about the score. Pomeroy’s rankings aren’t perfect, but I think it has Northwestern exactly where it should be for 2016-17. The Massey Composite Rankings, which take into account all Internet ranking systems, has Northwestern’s mean ranking at 31 as well.

For this particular Big Ten team, Ken Pomeroy has it spot on. Don’t tell Maryland fans that he has something right though.