- Henry Abbott, TrueHoop, NBA
- 0 Shares
You can lie with statistics.
But not like this.
What I mean is, let's say we're arguing about ... any old thing. Maybe you think people should eat more jello. No doubt you could find some stat (or, indeed, a stats-based rap) showing folks are beyond happy when they eat that stuff. Then maybe I could bust out metrics that favor pizza or ice cream.
And then we'd all just throw our hands in the air and say "THERE'S A STAT FOR EVERYTHING." Then we'd go home believing whatever we believed before we ever brought stats into it.
That's how life goes.
Except when it doesn't.
Every once in a while, you trip into something where, after you look at it every which way, the evidence really points in just one direction.
That's how we moved on from that whole "earth is flat" deal. It was once a real fight, of course. The stat geeks of their day talked about the lengths of shadows and the particulars of eclipses. Others trumped all that by saying, irrefutably "duh, it totally looks flat." Now we can see the darned thing from outer space, and guess what! It's round! All those theories of shadow lengths, they have graduated from theory to truth. Whether you're into statistics, science, math and the like, the planet you live on is shaped like a ball.
Along those lines normal stuff like saving for retirement, eating vegetables and exercising regularly have been shown by one sophisticated study after another to simply make sense.
You really can't win an argument against vegetables. Not now. I mean, yes, every issue technically does have statistics on both sides. But on a few special issues like these, the fight is just so vastly unfair. The pro-vegetable statistics are a thousand different indicators from a thousand different kinds of smart people reaching similar conclusions. You can lie with statistics, but not while making the kinds of cases people are making for vegetables. They are basically miracle cures, or preventative medicine, for darn near anything you can imagine. And the statistics on the other side have been cherry-picked by jokers and goofballs.
As new breed analytics have come to hoops, that's the stuff that excites me. The stuff where not only the best evidence, with the biggest samples, sifted through by the fairest-minded researchers tells similar stories again and again, but where it also lines up nicely (like vegetables and exercise) with things wise people have pretty much always known.
In other words, you can keep your petty little squabbles, the he-said she-said tug-of-war between spreadsheet-based and eye-test or tradition-based analysis. The far more fundamental and interesting question to me is: What is basketball common sense now? When you pan for gold in the muddy new soup of everything we know now from stats, video, coaches, players, GMs ... what glitters? What are the emerging "earth is round" things that, decades from now, we'll hardly fight about anymore, 'cause time will make those things look even smarter?
That's the good stuff. There are various pieces of it around, and I like the idea of collecting them. Some favorites, to get the list going:
Managing minutes works
More to come on this from our Working Bodies series as the season rolls on. This one is deep. But there are a few different worlds of evidence colliding here.
No team has won a title in a decade with a top player playing more than 3,000 minutes, even though during that period a who's who of MVP candidates has tried.
The Spurs blatantly lead the league in managing minutes ... and keep defying experts by getting more from aging players than could be expected.
Sports science is increasingly thick with studies showing athletes get injured less, and perform better, with workloads that are reduced from what NBA stars endure. Tweaking minutes in deference to that science has led to dramatic injury reductions, and performance improvements, in other professional sports.
Shoot more 3s
3s help a ton. The research started as a curio from John Hollinger that just attempting lots of 3s predicted wins. Now it's coming from any number of clever analyses. Shoot more 3s, win more. This theory has even passed a test in the field: The Rockets coached their team that way, and their offense got better. If you're not convinced yet, wait until SportVu optical tracking data makes it exceptionally clear over the next few years. It's not that every possession should end in a 3. But it's that NBA teams have been too careful with them, and that any coach who talks about "good shots" as distinct from 3s is wrong. Those are great shots.
Open shooters should get the ball
Here's another: Open shots are good. Really good. Way better than covered shots, as a group. The evidence is anywhere you look for it. A couple of years ago Ryan Feldman of ESPN Stats & Info. charted a bunch of crunch-time performances for TrueHoop, and found open role players vastly outperformed covered superstars. That was an honest, but simple look. Any number of more intense analyses have found similarly. There are great players and so-so players in the NBA, but the differences between them are seldom vast enough that you'd ever pick a covered guy over a wide open one. At this level, an open shooter is gold. And here's some recent affirmation, with a peek into the Rockets' famously private team of analysts. What is that geeky team obsessing over? Assistant coach Kelvin Sampson just answered a question about that from Rockets.com's Jason Friedman:
I’ll give you a good example: challenging shots -- we like to be around 70 percent shot challenges per game. I’ll get updated stats during the game about where we are in that category. Usually when you don't contest a shot they go in. When (Golden State head coach) Mark Jackson used to do games on TV, he would always say, "Hand down, man down." There’s a reason why you get a hand up. Well, we chart that.
"Usually when you don't contest a shot they go in." That's a big thing to say, and a big thing to know. It makes surprisingly clear what matters at both ends, and it means given a choice you'd sure like to shoot those uncontested ones.
We got confused for a while there. Every high-school coach knows all about the value of open shots, but I think we all kind of wondered is that true even if it means taking shots away from the best scorers on the planet, guys like Kevin Durant, LeBron James and Kobe Bryant? And the answer is, yes, it's true even then. If a shooter is in position and open, that's almost always a better option than a covered guy.
That used to be something we theorized, but it's getting, now, to be something we know.
You can lie with statistics.But not like this.What I mean is, let's say we're arguing about ... any old thing. Maybe you think people should eat more jello.