Iowa's offensive obviously outstanding, but it's the play of their defense that could make or break this season. So let's talk about that defense. I've been keeping tabs on how Iowa's defense performs when they play zone versus man-to-man defense as well as how they perform in transition. Here;s how Iowa has performed defensively in each of those scenarios when playing high major competition this season, as well as in each of the previous two seasons:
So this begs the question: why isn't Iowa running zone all the time? If Iowa's set half-court defense only allows 0.94 points per possession, then they're going to be Big Ten champions. So why bothering with man-to-man defense at all? The key question is whether that zone would still allow 0.94 PPP if Iowa used it all the time. Perhaps it is only effective in small doses and, over time, teams would figure out how to beat it. Those of you who have played a lot of basketball know that going up against a zone feels weird when you're used to facing man-to-man defense -- the defenders aren't where you're used to them being. It can take a while to get used to it. There are also systematic weaknesses with Iowa's zone that can be exploited. For example, over the years I've noticed that teams can overload one side of the floor and get an open corner three with Iowa's center having to attempt a very long closeout.
So how effective is the zone after it's been heavily used? To explore this, I graphed how this season’s zone defense has fared as a function of the number of zone possessions that Iowa has run in the game up to that point (shown as rolling averaged):
As you can see, Iowa's zone is highly effective for about 30 possessions and then it, uh, well it gets smoked. So far this season, when Iowa uses the zone for 30 or fewer possessions, they allow 0.83 PPP on those possessions. However, Iowa allows 1.17 PPP on all zone possessions after 30 possessions. I suspect this is why Iowa doesn't run zone exclusively. It is important to note, however, that Iowa has only exceeded 30 zone possessions in four high major games this season (Minnesota, ISU, @Rutgers, and @Maryland). So it's not like there is ironclad statistical evidence that using more zone would be doomed to fail.
On that note, I performed the same analysis for Iowa's games in the past two seasons and found no difference between the first 30 zone possessions and the subsequent zone possessions (0.99 vs. 1.00 PPP). The fact that previous years' zone defenses did not experience this calamitous drop in effectiveness makes me suspect that it is actually a statistical fluke. The bottom line is that I think Iowa should explore ways to use more zone. So far this season, Iowa has run man-to-man defense on 432 possessions and zone defense on 342 possessions (56%-44%). I'd prefer to see Iowa running closer to 60% zone to 40% man-to-man.
The effectiveness of Iowa's man-to-man defense also appears to declines as it's used more often. That can be appreciated in the following graph:
Iowa's opponents only score 0.92 PPP in their first 30 possessions against man-to-man defense; however, Iowa's defensive efficiency spikes to 1.2 PPP after that.
Since both Iowa's man-to-man and our zone defenses have been pretty effective in small doses, I think it makes logical sense to switch between them more frequently than Fran currently does. One thing that's become abundantly obvious through my possession tracking is that Fran does not leave the blackjack table when he is on a hot streak: if a particular defense is effective, he stays with it until the other team scores on 4-5 consecutive possessions, forcing him to adjust. This is a very logical system for Fran to have. However, I think that Iowa might be better to mix things up a bit more (i.e., a few possessions here and there), even when things are going well, to try and prevent the other team from getting comfortable against whatever defense Iowa is running. I'll continue to track the possession data to see how things progress over the rest of the season.