clock menu more-arrow no yes mobile

Filed under:

Scattered Scatterplots

Every few years, Ken Pomeroy produces a chart that tells you something that seems simultaneously counterintuitive and plainly obvious. Four years ago, it was a chart that demonstrated that 3-point attempts are generally a much better bet than mid-range jumpers (and layups are where it's really at, of course). Now, it's this:


and this:


Three-point shooting is a crapshoot. Particularly on defense, where teams appear to have about as much influence as they do over their opponents' free throw shooting percentage.

Pomeroy's take on this phenomenon:

Oh dear. The defensive plot is just a random scattering of data, as has been discussed previously, but the offensive version isn’t much better. If you shot 45% in the first half of the 2011 conference season, you’d be expected to shoot about 35% in the second half. If you shot 25% in the first half, you’d be expected to shoot 33% in the second half. A difference you couldn’t notice with your eyes. I don’t know exactly what implications this has on strategy, but when evenly-matched teams get together, action happening beyond the 3-point line is like a lottery. You take a shot and a third of the time you have success.

This observation should give us a moment of pause. Part of Michigan State's success this season has come from limiting opponents to just 28.2% shooting from beyond the arc--the fifth lowest rate in the country. In conference play, the number has actually been lower--26.8%, best in the league. The situation seems ripe for reversion to the mean.

While I'd bet on opponents shooting above 28.2% from here on out, I do think there are several reasons not to worry excessively about Lady Chance coming back to bite us.

Reason #1: Forcing tough three-point looks is what Tom Izzo's defense is designed to do (despite Izzo being a strict man-to-man guy). As Pomeroy points out, the percentage of shots a defense forces/allows its opponents to take from three-point range is a much more stable number than defensive three-point shooting percentage. MSU's opponents are launching from deep on 36.5% of their field goal attempts, which is right in the range the best Izzo-led defense have been in. Yes, life is random, but Tom Izzo has done a pretty good job beating the odds in the past.

Reason #2: MSU does some other things on defense pretty well, too. They're holding opponents to 42.4% shooting on two-point attempts, the 15th best mark in the country. And they're cleaning the defensive glass like a vintage Izzo team (26.6%, which ranks 16th; more on this below). This is a defense that, to date, has excelled both near and far from the hoop.

Reason #3: It's better to have held your opponents to a 28.2% three-point shooting than, say 36.3% (last year's number). If you're worrying about whether something that's gone really well might go not quite as well going forward, you're in good shape.

Non-scatterplot-driven thoughts on a three-point shooting number that is pretty clearly not purely a function of randomness. Keith Appling has only made 3 of his last 31 three-point attempts. His jumpshot is flat broke right now, which is now affecting the rest of the game. He's only made six total field goals in his last three games, and attempted just two shots from the field Saturday night. Hopefully, a more up-and-down game against Indiana will do him good. And he's continued to get opportunities to score at the free throw line (some of it at the end of games). But fixing the mechanical issues with his shot (it's not just mental--a lot of his shots are barely even getting to the rim) is, I'm sure, a major priority for the coaching staff right now. Otherwise, defenses are going to continue to sag off him and reduce good shooting/penetrating opportunities for the rest of the perimeter players.

Back to scatterplotting non-correlated numbers! As we move into March, you're going to be hearing a lot from national commentator types about MSU's stellar rebounding numbers. And rightly so. But, keep in mind, every time someone sites rebounding margin, they're actually shortchanging the Spartans. Offensive and defensive rebounding are two different things:


This has been true in prior seasons, as well. Teams that rebound well on one end of the floor aren't all that much more likely to rebound well on the other end. As Con-T pointed out, everybody tries to rebound on defense; offensive rebounding is much more a function of strategy (Bo Ryan says, thanks but no thanks). Even among those teams that clearly pursue second chances on offense, though, good defensive rebounding does not automatically follow. Of the top 50 teams in the country in offensive rebounding percentage, only 6 also rank in the top 50 on the defensive end. And MSU is the only team that currently ranks in the top 20 on both ends (see the arrow in the chart above).

So word to your commentator: When it comes to rebounding, MSU is good at two things, not one thing. (Additional Con-T observation: The individual numbers further emphasize this notion. It's not the same guys driving the rebounding numbers on the two ends of the court. Draymond Green is devouring what seems like ALL THE DEFENSIVE REBOUNDS, while Branden Dawson--and, to a lesser extent, Derrick Nix and Adreian Payne--are earning their keep on the offensive glass.)

And one more glorious scatterplot for the road. It's been a while since I've seen Big Ten efficiency stats displayed graphically. This is the time for that to end:


[student section chanting] WE'RE MOST EFFICIENT, WE'RE MOST EFFICIENT [/student section chanting]

There's plenty of season left here (somewhere between 4 and 11 games; give me 8 and I'll move on to baseball season quite contentedly), but this has already been one for the books--from preseason unranked to statistically-dominant behemoth. This is the closest thing we'll ever get to 1997-98 again. Enjoy it.