The defense is weird and somewhat incomprehensible this season. On paper, it looks like they're doing exactly what we've been begging for over the past several years: packing it in more, extending to the arc less, staying at home without gambling to protect the rim and box out.
Top 100 in 2pt% allowed for the first time since 2011. Top 50 block% for the first time since 2006. 4th best defensive rebounding% in Kenpom history. Sub-100 in 3pt% allowed for only the 4th time in Kenpom era. One of the very worst defensive turnover rates in the country. It all points to a more conservative defense that tries to be fundamentally sound without opening itself up to breakdowns with missed rotations, rather than the usual idealistic defense that aims for high turnovers and high rim protection and high 3pt prevention. Most strikingly, in Kenpom era, the worst Duke has finished in 3pt attempt rate allowed is #24. This Duke team is currently #181.
The justification to still hate K and his defensive scheme this season is that these numbers are all heavily skewed by the amount of zone Duke has played, so the man to man defense is still probably K's overextending garbage, but without generating turnovers at all. I don't know what the breakdown of zone vs. man to man possessions has been, but Kenpom seems to agree with this to some extent, since Duke's defensive fingerprint is "Inconclusive" for the first time ever. It has always been "Mostly Man" before this season.
Also throwing all of this off is Duke being #265 in FG% allowed at the rim on hoop-math. So maybe they're not protecting the rim at all, or maybe they're breaking down so badly that a lot of shots at the rim are uncontested layups, as would be normal for a recent Duke defense. Duke is crushing non-layup 2-pointers, at #23 FG% allowed on those shots.
I have no conclusions from this. Seems like more of a work in progress than any other Duke season. No identity yet. IMO, this is preferable to having an identity of giving up layups.