Ken Pomeroy had a fascinating series of posts on his blog (Note: this is not behind the paywall, so everyone can read them) examining the affect of defense on 3-point percentage. It turns out that, contrary to what we would expect, there seems to be no relationship at all between defense and 3-point percentage. Please pay attention to the word "percentage," because there are other factors the defense does have an influence on, like 3-point attempts. The better the defense, the fewer 3-point attempts are generally made, even though the actual percentage of made threes taken appears to be independent of the defensive effort.
Let's take the isolated example of the Vanderbilt Commodores vs. Kentucky Wildcats on Saturday. On the season, Vandy averages about 41% of their shots from three. You would expect that to vary depending on the defense, and you would be right. Here's what Pomeroy has to say about that:
3PA% [3-point attempt percent] is one of the most predictable shooting-related stats around. We’ve already established the defense has influence over this, but a major part of an offense’s style is how often they want to shoot threes, so they have influence, too. More influence than the defense, actually.
What is left unsaid here is implicit impact of coaching. Coaches try to convince players not to take defended threes almost universally (except in the case of, say, a John Jenkins, who is such a good shooter the defense has to be exceptional to make him miss). That means that good 3-point defense should result in fewer three point attempts due to a combination of coaching and player basketball IQ.
So did Kentucky succeed in forcing Vanderbilt to shoot fewer threes than they normally do? On Saturday, Vandy took 63 total shots, of which 18 were threes, which works out to about 29% of their shots. Back during the game in Nashville, Vanderbilt took 63 shots again, 21 of which were threes for a 3PA% of 33%. So from that, it's fairly easy to conclude that Kentucky accomplished the reasonable objective of making Vandy take fewer 3-point shots than they normally do, since neither of those approaches Vandy's average of 41% 3PA%.
But did that great defense lower Vandy's 3-point percentage? In game 1, the Commodores shot 38.1%, and Saturday, they shot 38.9%. That is not appreciably different from their season average of 39.8% -- basically, what they do to everybody, on average. This is just anecdotal, of course, and included mainly for the sake of interest. Kentucky's excellent defense only managed to hold them to their average 3-point percentage.
In the end, after 3 posts, numerous simulations and a partridge in a pear tree, Pomeroy came to this conclusion:
I think the takeaway from this week is that 3P% is largely out of direct control of the defense, and it’s a lot less predictable than one might think when two teams of reasonably equal ability face each other. When another team makes a surprising run in the NCAA or even its conference tournament people are going to wonder why we didn’t see it coming. However, there are some things that you can’t see coming and 3-point shooting is one of the best examples of that.
Does that make you a little uncomfortable about Kentucky's chances? After all, he is saying that just about anyone can go all Sam Houston State on UK (or anyone else), drop 18 threes on them and send them packing early.
But don't let it worry you -- I know I don't. Teams that go nuts from three can certainly take a game that should be uncompetitive and make it very interesting, but the very best response to a 3-point onslaught, overall, is offensive efficiency. In the SHSU example from above, the Bearkats made 18 threes at 47% 3P%, but UK still had a much higher eFG%, 69%-49%, and a much higher offensive efficiency, primarily due to the size and talent advantages they had.
The reality is, if a similarly-talented team goes nuts from 3-point range, it's hard to overcome. But if an inferior team does so, your odds are much better, especially if you are the second-most efficient team in the land offensively.