No one thinks this year's version of the Detroit Pistons is particularly good at defense, at least not so far. One particular area of concern is opponents' 3-point shooting, where night after night, the Pistons leave shooters open and are punished.
And while there are two schools of thought on DBB as to why this happens (these are, of course, the "it's all Josh Smith's fault" school and the "it's only mostly Josh Smith's fault" school), it occurred to me to wonder what's really going on here. Are we truly playing defense at a historically bad level, or is it possible we've just had the misfortune to face a few hot-shooting teams so far this year? After all, when our shooters get a wide-open look it's not like the ball automatically goes through the net, so if it feels like the other team is hitting all of their wide-open looks then something else is probably going on.
So I took a look at the numbers, and found a few surprising things. First, it turns out the Pistons are not the worst team in history at 3-point defense, or even one of the worst this year. The Pistons turn out to be 10th-worst in this category, giving up 3-pointers at a .366 clip, compared to a league average of .359. Gosh, maybe this isn't even a huge area of concern! But then after tonight's awful performance, which drove our defensive percentage up to .370 (7th-worst in the league), I figured it was worth going a little deeper.
Mainly I wanted to look at who we've played. After all, like most Eastern teams, we haven't exactly been playing a league-average schedule. Our strength of schedule only ranks 21st, and we've played a specific subset of the teams in the league. So have we possibly been playing teams that are particularly good at shooting 3's, or maybe particularly bad?
Here's what I did. For each of the 22 games we've played so far, I looked up how many 3-point attempts our opponents had that night. Rather than caring about how well they actually shot on that given night, I looked up their average 3-point percentage for the season - in other words, how well they have shot against a generic team. And then by multiplying those two numbers together, I find out how many made 3-pointers we "should have" given up on that night, if we had played defense like a generic team instead of like ourselves.
What I found was this. If every team had shot their season percentage against us this year, we would have given up 179 made 3-pointers in our 22 games so far. And what actually happened? Turns out we've surrendered a total of 181 3-pointers in the real world. By this measure, we're almost exactly where we should be.
Or, to look at it another way, opponents have shot .370 on 3-pointers against us this year. But the weighed average of the specific opponents we've played is actually .368. According to the numbers, we're defending 3-pointers almost exactly like an average, generic team; it just so happens that we've played some very good 3-point shooting teams so far (like Golden State, Portland, and LAL, all of whom are scorching the nets at a .400 percentage or better against the entire league).
(Parenthetically, looking at our opponents' averages you'd expect them to have taken 463 3-pointers against us this year. In fact they've taken 489 of them, which is more than 1 extra attempt per game that we're allowing, although teams aren't converting those extra attempts any more frequently than they usually do. I'm not sure if these numbers mean anything.)
So anyway, does this all mean that our 3-point defense is just fine? Well, I wouldn't go that far. Luck plays a big role in determining whether an opposing player knocks down an open 3-pointer, and over small sample sizes that luck can lead to huge variance in results. When Joe Johnson goes 8-for-10 from behind the arc, as he did against the Pistons this year, it's obviously not just a function of being left open, because you rarely see anyone shoot that well in a game no matter how bad the defense is. 8-for-10 has to involve some bad luck on our part, or some really hot shooting on their part, depending what you want to call it. So it's still possible that our 3-point defense is worse than normal, and we've actually been fortunate thus far that we haven't surrendered more makes, although I doubt any of us feels lucky.
As a concluding point, the luck factor is why some people who are smarter than me about stats have questioned whether 3-point defense is even a real thing that teams can actually be good at. Statheads have looked at both the college and pro games and found that in addition to luck being a big factor in 3-point defense, being good at defending the 3 is largely correlated with having a good defense in general, rather than being a specialized skill. (Common sense: if you're good at defending down low, teams will be more eager to look for the 3-pointer, hoisting up some shots they wouldn't take otherwise.) As Ken Pomeroy summed it up:
- The offense is largely in control of the quality of 3-point shots it takes.
- These decisions are affected by the quality of the opposing 2P% defense
- 3P% is also influenced by effective challenging of shots.
- All of that can add up to about a 3% swing from average.
- So 3P% defense is not totally random
- But a defense has considerably more direct impact on 2P% than 3P%.
There's definitely plenty to be concerned about when it comes to the Pistons' defense. But it's looking, counterintuitively, like 3-point defense isn't actually a specific area of concern. Teams are basically shooting their normal percentage against us from 3-point range, and there's relatively little we can do to alter our opponents' 3-point percentage directly. Instead, we can work on beefing up the defense as a whole, and more misses from 3 should ensue as a natural consequence of that.