FanPost

Derrick Williams and the Fiscal Cliff

I recently had lunch with my dad for the occasion of celebrating his recent retirement after 42 years of working. We ate delicious Himalayan food and talked about what he'd been up to over the recent number of days since retirement. He recounted to me how one morning he had some errands to run at a local grocery store, and while he was there he noticed signs up advertising that his local congressman was going to be there for a meet and greet. In fact, the congressman was going to be there 15 minutes from the time that my dad was reading the sign. Normally my dad would never be able to just hang out for such things, but now that he was retired it occurred to him that not only COULD he hang out, but in fact he SHOULD hang out. As he put it, 'How else is my representative supposed to know what I want if I don't tell him?'

One of three things my dad told him that day was that he was tired of people using the phrase 'fiscal cliff.' It's not a cliff, and the implication of the phrase suggests that things will change immediately and irrevocably on January 1st if politicians don't do something. It's sensationalising economics, and my dad is tired of it. He believes it obfuscates the real work that needs to be done because it keeps people in polarized positions - either you're against the fiscal cliff (which of course means x, y, or z position), or you're for the fiscal cliff (which means you don't aspire to a long career in politics). Either the way, the phrase connotes an expectation that is misleading and (even worse) directs the conversation about it down the wrong rabbit hole(s).

So what in the hell does this have to do with Derrick Williams?

Derrick Williams is currently the second best rebounder on the Timberwolves. Better than Pek. Better than Dante. He's also the second best three point shooter (which kind of undermines a point I was going to make, but whatever). In fact, he does a lot of things pretty solidly except display the kind of offensive awareness you'd like to see in knowing when to shoot (and from where) and when not to.

Coming into the draft the hype about Williams (that I remember) was two things - he was athletic (dude could finish with authority) and he could shoot (57% from three!!!!!). The reality of his NBA career after essentially one full season (78 games) is that he's been terrifically underwhelming. His three point shooting sucks, he often seems lost offensively, and he can't even beat out Dante Cunningham on his own team! A guy who was a second rounder!

And to a degree all of this is true.

And misleading.

The fiscal cliff is to the economy what getting taken #2 overall with 57% three point shooting is to Derrick Williams: a misleading set of expectations that sensationalizes the reality of is before you.

I need to get up early tomorrow, so I'm going to blast through the essence of my data pretty quickly, trusting that if you actually want to argue differences you'll take the time to look at and reinterpret the data yourself. Mostly I'm drawing from bball ref (the link above), 82games, DraftExpress, and Hoopdata. I promise that I'll do my best to not cherry pick points that help a certain point of view, because really what started this was looking at bball-ref to compare DC with Derrick, expecting that DC was going to be a superior rebounder to Willams. And he isn't.

Not per36, not per game, not even by TRB% or DREB%. The only metric by which Dante is a better rebounder this year than Derrick is on the offensive boards. I was not expecting that at all. Here's the head to head comparison in case you'd like to see it.

It got me thinking back a little on what I've seen from Derrick this year, besides the bad shots, and looking more closely at the stats combined with what I've seen made me realize that I've been looking at Derrick under false expectations. He's a decent three point shooter (college and pro combined he's .347) but a better than expected rebounder and (until Adelman stopped playing him) an improving defender. He gets to the foul line pretty well still. His biggest problem is that he simply doesn't know when to shoot and when not to, and compared to Dante Cunningham it's a glaring weakness.

Williams is second on the team in the rate of shots he jacks up per minute, yet he's way under .400 for FG%. If you compare this year's close to the rim percentages to last year's, well, you begin to get the idea of what's happening. The last piece of the puzzle is that while he's making (so far) more threes while taking less, he's also added more than 1.5 shots per40 from the worst possible spot on the floor - the long two.

The return of Rubio and simple regression to the mean suggests that his tips and dunks will begin to fall at some point. He's not going to go the whole season shooting .250 from dunk, er, dunks...whatever, unless Adelman sits him the rest of the year. He was 90+% last year. He's rebounding with authority when he's out there, his defense is improving. Let's call it the Kirilenko effect, and let's hope that it includes one last facet - when to shoot and when not to.

I'll finish by saying that I expected to make a good argument that Derrick is a bad three point shooter. But the truth, as born out by the numbers, is that he's mediocre, which is not (by definition) bad. He's about average, and that's OK. What's interesting is that he's at about where he should be given his career (college and pro) numbers. Outside of his one good year at AZ shooting from downtown, he still shot .300 from distance. Again, not great, but not turrible either. Of course it's very tempting to say that .570 from three is a huge fluke, so I decided to run some statistical analysis on it.

Using his overall career (college and pro) three point shooting MINUS his one good season (I used 3FG and 3FGA to calculate the percentages), I modeled 1000 seasons to see how likely - or unlikely - a season of .570 would be. It's important to understand at this point what the purpose of this modeling is. Essentially we're going to find one of two results - either it's pretty likely that such a season will occur (given 1000 seasons), or it will be quite unlikely that it will occur. One proves that the initial assumption is correct (that he's the three point shooter of the three season outside of his one good season), and the other proves just the opposite - that we can't assume that the shooter of those three seasons is the 'norm' for him BECAUSE the one good season was such a statistical outlier that there has to be more to the story. It wasn't produced by random chance, so what produced it?

Ready to hear the results? 1000 seasons modeled and his one good season bested the outermost extreme outlier by .100 points. Can't really tell you much more about what it means other than it's highly unlikely that a .300 3fg% shooter could ever have a stretch of converting around 100 3PA's at a .500 clip. It's just about impossible.

So - Derrick Williams. I'm not suggesting that he's a future all-star, or that he's even long for this team, but I am suggesting that other around the league might look at this numbers (offensively and defensively) and see that he appears to respond to coaching and might see more value in him than we do, or at least still see value in him. Per36 he's a 16 and 10 player right now, today. Now how is that not worth something?