An Analytics Analysis of 2's vs. 3's

Self has recently averred that this team needs to shoot more 3 pointers. But in the past he has been averse to depending too much on 3 pointers, even calling it fools gold. So why the change. The answer is simple, analytics. On the surface it seems that since a 2 pointer is worth 2/3rds of a 3-pointer that you only need shoot a percentage of 3-pointers at 2/3rds the rate of 2 pointers. This is far from true. And the complete analytics goes beyond just a comparison of the shooting percentages. But first, lets look at the shooting % comparison and get that straight before we bring in some other analysis.

There are 4 key variable we will need to make this calculation of what is the equivalent shooting percentage between 2's and 3's. These variables are;

2 point shooting %
3 point shooting %
Offensive rebound % off of missed 2 point shots
Offensive rebound % off of missed 3 point shots
% of possessions a team turns over the ball (%LoB)

Now let's take the NCAA averages for these stats (using last season's totals)

2 point shooting % --- 49.92%
3 point shooting % --- 33.47%
Offensive rebound % off of missed 2 point shots --- 40%
Offensive rebound % off of missed 3 point shots --- 20%
% of possessions a team turns over the ball (%LoB) --- 19.0%

Using these metrics, we can calculate the points per possession assuming 2 point shots and 3 point shots. this is as follows;

PPP 2-pointers --- 1.191
PPP 3-pointers --- 1.062

So clearly, conventional wisdom says that most teams last season who relied on 2-point shots were better off than those who relied on 3 pointers. So why all the emphasis on 3 pointers? Well again, we will leave out for no the strategy of crashing the offensive boards versus dropping back on defense and stick with the overall averages of offensive rebounds used above. You can see that the big issue in this calculation is that the average team grabs the offensive rebound twice as often on a 2 point shot as on a 3 point shot. This is the absolute key to understanding how a team develops a strategy.

First, let's figure out what a n average team would have to shoot from on 3-pointers to make the PPP equal to 2 pointers. Assuming a team shoots 50% on 2-pointers, they would have to shoot around 37.3% on 3-pointers to get the same points per possession. Roughly speaking, the conventional wisdom is that a team needs to shoot at roughly 75% from the 3 point line as they do on 2 pointers to be equivalents in points per possession.

So last season Kansas had a fella named Udoka Azubuike manning the middle. Thanks in large part to him KU made 55.3% of their 2-pointers and 34.4% of their 3pointers. Plus (and this is the real amazing part) KU grabbed 46% of the offensive rebounds on missed 2 pointers and 21% of the offensive rebounds on missed 3-pointers. So lets plug these numbers in and see where KU's PPP land on 2 pointers vs. 3 pointers. KU scored 1.40 points per possession on 2 point shots and only 1.10 points per possession on 3 pointers. That is monstrously disparate. KU should have looked for a 2 pointer every time down the court and they mostly did. It was a wildly efficient game plan thanks to having a guy like Udoka in the middle. So again, why all the focus on 3 pointers?

Let's look at 2021 thus far. Through 9 games, KU is shooting 39.9% from the 3 point line and 46.5% from 2. More amazingly, KU his grabbing 46% of offensive rebounds on 2-pointers and an incredible 27% of offensive rebounds on 3 pointers. So let's do the points per possession for these numbers.

2020-21
PPP 2-pointers --- 1.125
PPP 3-pointers --- 1.311

What? KU is far better shooting 3 pointers this season than 2 pointers. Is there any wonder Self has all of the sudden become a fan of 3-pointers. Let's read between the line of something else Self has been saying lately. When asked about the expectations of David McCormack, he has changed his tune lately to saying things like "there is a lot a player can do besides score to have a huge impact on the game. Let's not put unnecessary expectations on David, but let's put him in situations where he can use his size and athleticism to impact the game." Now let me give you my interpretation.

KU opened the season with a game plan similar to every other year under Self. There were expectations that McCormack would fill that position that is critical in the Self game plan. Namely, a guy who can consistently make 60%+ of his shots and rebound 40%+ of those misses. In other words, under Self, KU has typically been around 1.35 points per possession on 2 pointers and that is the money game. But 5 or 6 games in, Self was getting a clear sense that David was not going to be that guy, at least not anytime soon and possibly never. And it was not for lack of trying. It was clear from the focus on getting McCormack shots that Self was focused developing this money game of a highly efficient 2 point game yielding 1.35 PPP or more. But even against weaker competition, McCormack was not showing the aptitude or inclination to be that guy. Meanwhile, back at the ranch, the perimeter guys were showing a great ability to hit three pointers and more importantly shoot intelligent 3 pointers. We don't have a cowboy who wants to jack up threes and who is likely to go 1-10 in a game. Rather we have 3 guys who can all shoot for percentage and a 4th (Garrett) who is showing he can hit the open 3. So Self saw that we are very likely to be that team that hits more than 75% from 3 than we do from 2. And to add a further icing on the shooting cake, this KU team has the best perimeter rebounders we have had collectively since 2008. We have had great perimeter rebounders, but the combination of Wilson, Braun and Garrett is the best combo and Agbaji is being influenced by these guys and is getting in the act. So we are rebounding missed 3 pointers at a prodigious rate of 27%.

I do not believe for a second that Bill Self is changing his philosophy of the high-low and recruiting more perimeter players. KU was the best team by a fair amount last season and it was largely because we had an incredibly efficient and effective 2 point shooting attack. If Self could find a Udoka clone, we would be right back to that game plan in a heartbeat. It is the money game. You can always lose to a hot shooting team on any given night, but you will win more than you lose with an effective 2 point game. But the key is, you have to be flexible, because a well-coached team with a great game plan can always counter your best weapons. One of the reason that Coach Self has won more games than any other coach in the last 17 years is largely because he has consistently used the money game (think Simien, Arthur, Aldrich, Withey, Robinson, Azubuike) but also because he has been better able to adapt to game plans specifically aimed and scouted at stopping KU's strengths. What Self is doing is 2020-21 is adapting to what his team has showed him they are capable of. The numbers this season are clear as glass, SHOOT MORE 3's!! Self is not changing his stripes. He is simply being the great coach he is and using the best players he has to develop the best game plan for success.

Having said all that, I can almost guarantee you that in practice, he is still working McCormack's butt off to get him to be a more effective post player in the mold of Self's vision. But in the mean time, Self is going to develop the most effective game plan for the personnel he has.

One more thing. I mentioned that the analytics involved more than just the 5 variables mentioned above. These are things such as the philosophy on crashing offensive boards vs. hustling back on defense and how that varies on 2 pointers vs 3 pointers. Another important factor is what you do with offensive rebounds. If you have watched Jay Bilas much you have heard him espouse the philosophy that the best 3 point shots come off of offensive rebounds. That is a very important part of a game plan. Well-coached teams are trained to where the perimeter guys who are not crashing the boards are trained to spread the court and find an open place on the arc so if they grab the offensive board, the rebounder is trained to immediately look to get the ball to one of those open shooters. Even if a defender recovers, a 2nd pass can be made to another open shooter. And statistically speaking, this plan is geared to offensive rebounds from close-in 2 pointers, further enhancing that the efficacy of the first shot being a 2 pointer. This was a huge focus of KU last season. A good percentage of our 3 pointers came off misses of 2 pointers and offensive rebounds. Statistically, it is much smarter for perimeter guys to hustle back on defense on a missed 3 pointer since there is twice the probability the opponent will get the rebound. So how a coach coaches to crashing the boards on 2 pointers vs 3 pointers is another important metric to determining a 1st shot game plan. There is more, but I will leave it here for now since this has covered the most important analytics of how a coach will develop a 1st shot offense system or game plan.

Great post Corpus. KC Star touched on same this morning without near the detail. It is interesting to watch what the players do with the ball once received on the perimeter after an O rebound. If not shooting the 3, Wilson, Garret and Braun tend to drive paint to try and score more than Obagi who seems more likely to reset offense. Have noticed Braun using ball fake pass to corner to free up shot or drive. Would love to Obagi develop more of a handle to do some of the same.