Recently, when Glenn looked at the early season trends of the Wildcats, one of the areas he mentioned caught my attention. Glenn lamented the low assist rate the Cats are currently putting up and I've read comments over the last couple weeks by other people who feel the same way, though mostly phrased in terms of wanting to see Knight average more assists per game.
The assist is an interesting number. It's the only statistic in basketball that directly measures an element of interaction between two teammates. One player can not earn an assist unless another player makes a field goal. We tend to treat all assists the same, but if you stop and think about it we probably shouldn't.
Broadly speaking, most assists are earned in one of two ways: the first is by a player making a 3 PT field goal, the second is by a player making a shot around the basket (layup, dunk, bank shot, etc). In the first instance, most of the credit for the assist really ought to go to the player who made the shot - after all it's very easy to pass to a player on the perimeter, but it's more difficult to hit an outside shot. In the second case things are reversed: it's very easy to hit a shot near the basket, but it's very difficult to pass the ball to a player in the post in a place where he can actually do something with it.
With this in mind, my feeling has been that the lack of assists so far this season is due more to not having Patrick Patterson or DeMarcus Cousins around to throw the ball to than it does with "selfishness" or "poor decision making" or any of the other usual buzzwords we typically hear when talking about low assist numbers. To that end, I investigated (to the extent possible) how much the lack of a Cousins-esque or Patterson-ish inside player might be hurting that column in the box score.
What I've done is go through the play-by-play data for this season and last season and determine how many assists came from 3's and how many came from 2's. To gain some perspective and have something to compare with I also did the same for UK's opponents in these seasons. For last season I have both the full season numbers as well as those for the first 9 games, but as you'll see there's not a huge difference between the two.
|UK 2010-11||UK 2009-10 F9||UK 2009-10||OPP. 10-11||OPP 09-10 F9||OPP 09-10|
|3PT FGM assisted||66||51||211||31||49||184|
|2PT FGM assisted||45||89||362||68||54||234|
|2PT Assist %||27.27||44.28||43.35||40.96||41.04||37.14|
The first thing that struck me was how high the assist rate was for 3's. This year and last, players simply don't get many threes on their own. I have no idea if this is because they don't make many of these types of threes, or if it's because they don't attempt many to begin with. Either way, I expected this number to be high - I did not expect it to be this high. As you can see, the rate for UK's opponents is much lower, indicating that opponents are getting more threes on their own without the help of their teammates.
The second thing is the big dropoff in 2pt assist rate between last year and this year - nearly a 20 point difference. There is not a whole lot of difference between this year and last for opponents. You can see from the total numbers that opponents have made nearly the same number of field goals through the first nine games and have nearly the same number of assists, but more of those makes are coming from inside the arc than they were at this time last season.
That pretty strongly confirms what we believed: without a premier big man in the center to throw the ball to, most of Kentucky's inside production must come from the kinds of plays that don't result in assists, namely players starting with the ball away from the basket and either driving or backing their way to the hoop.
This brings us to Brandon Knight and his ongoing development in finding the proper balance between scoring and setting up teammates to score. It's easy enough to say Knight needs to get more assists or that he needs to make better decisions with the ball, but saying those things is a bit like when I tell my calculus students they need to study more to do better on tests - it's true but not terribly helpful. It makes more sense to figure out some specific goals or areas for Knight (and other guys) to work on and this suggests one such area.
Josh Harrellson has shown that while he's not Patterson or Cousins, he does have some ability inside. With his limited touches on offense Josh has been one of the most efficient scorers in the nation with an offensive efficiency of 136.5 and he makes good decisions with the ball as evidenced by his miniscule 13.6 turnover rate. If the primary ballhandlers (Knight, Lamb, Miller) were to make it a goal to find 2-3 possessions in each half where they specifically look to get Harrellson the ball in a position to score, I think we would see an increase in the assist rate as well as a few more points scored. Josh isn't going to convert every opportunity of course and depending on what the defense does he might not even attempt a shot and instead pass the ball back out to someone else. But he has shown some offensive ability and it makes for a straightforward way to improve the team by taking fuller advantage of it.