JacobPadilla wrote:SeattleJay wrote:Jacob, I assume we are better at FT differential rather than the number of attempts we take? That doesn’t help?
While Creighton leads the country in lowest foul rate, teams shoot at a nearly 75% clip when they do get to the line against CU (304th). And while they get to the line a lot more than their opponents (28.8% vs. 16.5%), the Jays still get there far below the average rate (33.2%; CU is 302nd in FT rate).
I don't have a firm understanding of how exactly Ken calculates his ratings and how things are weighted; just pointing out some of the outliers on their sheet and how they may impact the final net rating.
It literally doesn't take into account any of that stuff. Just straight net efficiency, not mattering in the least how you go about scoring and defending.
We dug a hole because we had three terrible performances (Nebraska, SDSU, Gtown) in which we underperformed expectation by an average of ~23 points per game. The Gtown game in particular was a performance on par with what you'd expect from the ~300th best team in the country. A loooot of teams above us don't have anything like that on their record (among other factors). To that end, you can go to Torvik and sort dates to see that from the NU to the Gtown game we were only the 75th best team in the country (despite the Kansas win). We would jump ~4 spots at Kenpom if that Gtown game was simply a close loss instead of a blowout. Its not just that we lost these games, its that our per-possession performances were atrocious.
On the flip side, we've been slow to climb the predictive rankings because we've only slightly overperformed expectations. While we've obviously played well and our resume metrics have improved considerably, we've only overperformed expectation in the last 5 games by about 4 point per game. Winning at Nova by 2 instead of the projected 1 is a great Q1 resume-building win, but its not going to shift the predictives at all because it almost perfectly matched expectation (ie. we've played only a little better than you'd expect the 30th best team to perform). By the same token, losing that game by 2 wouldn't have hurt the predictives either even if it would have been considerably worse for the resume.
Case in point - Torvik suggests we're the 19th "best" team (net efficiency) in the country since Jan 1 despite having the 9th most impressive collection of wins and losses. As a point of interest, we're basically the opposite of that 6 seed/elite eight CU from a couple years ago that had the predictives of a 3/4 seed but the resume of a 7/8 seed.
On the NET topic, because this is something people are often somewhat confused about - The NET is NOT a resume metric, and it NOT driven (primarily) by your wins and losses or your quad records. The NET is an efficiency metric, akin to Kenpom, used to assess the quality of your wins and losses. It is a predictive metric (Kenpom, Torvik) rather than a resume metric (Wins above bubble, strength of record). The NET informs the quad records, not the other way around. The "Ohio St. is this and that" argument is explained by their net efficiency margins, not their win/loss record. And thats fine - the NET is a sorting tool, not a ranking tool. A team's performance against the NET is far more important than their actual NET ranking. Which is why we'd be more like a 6 seed today than an 8 like the raw NET number might imply.
Hopefully I explained that in a way that makes sense
