clock menu more-arrow no yes mobile

Filed under:

Playing the Lottery: a statistical look at 3-pointers

Villanova's opponents have shot a lot of threes against them, Jordan Sperber takes a statistical look at why that is.

Howard Smith-USA TODAY Sports

We simply don't know just how good this Villanova team is quite yet. Being +85 four games in while off to a slow shooting start (25.5% from three) is probably a good sign, but the Battle 4 Atlantis is our first real look at high quality competition.

The focus early in the college basketball season is to find out as quickly as possible how good your team is, but this is actually only half of the story. Not only is team *quality* important, but so is team *consistency*. Consistency, or variability in statistical terms, is what makes March so much fun. Consistent teams aren't automatically good or bad. At 0-28 last year, Grambling St. may have been the most consistent team in the country. However, combine quality with consistency and you have a championship contender.

Last season, Villanova illustrated this concept of variability. On an average day, Villanova wouldn't have been able to beat Louisville or lose to Columbia.  This variability was arguably a good thing for a team in somewhat rebuild mode following a disappointing 2011-12 season. In 2013-14, expectations are no longer the same. It's very reasonable to expect an increase in quality, but what can we expect regarding variability?

Many things in basketball are sort of "predictably unpredictable", but none more notable than three-point shooting. Two seasons ago, Ken Pomeroy took a look at defending the three-point line in a must-read article for basketball fans. He looked at teams' 3P% allowed (on defense) in the first half of conference play and in the second half of conference play and found that there was remarkably no correlation between the two. The stat that did correlate from first half of conference play to second half of conference play was three-point attempt percentage. In other words, teams can control whether their opponent takes threes but not really whether they make threes.

41% of all shot attempts by Villanova opponents this season have been threes. That's good for 331 out of 351 D1 teams. Pomeroy described this style of defense as "playing the lottery". Villanova's interior defense has been great to start the season. Some nights the Wildcats will win the lottery playing their current style of play, but with a strong defense it would be nice to see Jay Wright stop buying tickets in the first place.

I took a look at Villanova's defense in every game since the 2010-11 season (a.k.a - post Scottie Reynolds). With these 102 games, I separated them into two groups. The first group is games where less than 32.5% (NCAA average) of opponent shot attempts were threes. The second group is games where more than 32.5% of opponent shot attempts were threes. On average, Villanova's defensive performance (adjusted for pace of game and strength of opponent) was virtually the same for both groups. However, take a look at the distributions below:


In the graphs above, the x-axis labels are relative to an NCAA average team. As you can see, Villanova is rarely below NCAA average defensively in any given game. Notice the larger distribution when the Wildcats give up a lot of three-point attempts. In the 63 games where this was the case, Villanova had seven games of extremes (very good/bad defense). This is exactly what "playing the lottery" means. In the other 39 games, Villanova's defense was much more consistent from game to game.

So now that we have a better idea of what happens when the Wildcats give up threes, the question becomes: Why are they giving up threes?

Villanova's team defense was good last year and has been just as good this season. The Wildcats have been able to force turnovers in the 1-2-2 press and defend the paint in the half-court. Going to the film, the three-pointers allowed have generally not been big defensive breakdowns. Sure, we can expect an occasional wide open three allowed after an unsuccessful press, but that's the price to pay for creating turnovers. Below is a video (with annotations) of two defensive possessions against Delaware that shed some light on the three-point situation:

In both clips, Chennault and Arcidiacono rotate correctly to the ball. All five players on the court actually rotate correctly, a very encouraging sign. However, forcing shooters off the three-point line simply doesn't seem to be a primary objective for the coaching staff. Jay Wright appears content with allowing semi-contested threes. This strategy is not necessarily wrong. Putting an emphasis on getting all the way out on shooters would almost certainly allow opponents to get into the paint more. This would be a good thing according to Villanova's current interior defense, but maybe the increase in penetration would drastically change things. This depends a lot on the ability of the defender closing out to react and slide effectively.

This style of defense has been consistent throughout Wright's tenure at Villanova. Even the 2009 team consistently played the three-point lottery on their way to a Final Four. That being said, expect some more dramatic highs and lows on the Main Line in 2014.