In the interests of keeping our team away from rote and boring features, we don't intend to do season previews this year. At least not officially. But I tried to do a few longform previews last season in the form of reflections -- for today's Capsule (Plus) on Kevin Durant, I'm going to attempt to work in a general preview to the Thunder's season. So let's see if I can make this work.
Last season, I did a variety of playoff predictions. Many were apt. For instance, I correctly predicted that the Lakers/Nuggets series would be a coin toss. Mission accomplished. I predicted that the Thunder would make short work of both the Mavericks and the winner of Lakers/Nuggets, whoever it may be. That's essentially exactly what happened. But there was one thing I believed very strongly during the duration of the preview season, something that ended up partially becoming my downfall as I turned out to be so wrong as to shock and appall me. I predicted -- with the utmost confidence -- that the Spurs would beat the Thunder, no matter where or when they match up. I predicted, in fact, that it wouldn't be close.
Let me state the obvious. I was completely, unfathomably, misguidedly wrong. Utterly and miserably. But we'll get to that later.
• • •
I've told many people over the past few months that in the rubble of last year's Western Conference Finals, I thought I'd descended into a blithely venomous antipathy for Kevin Durant. I don't really retract that, because it's still completely true. I barely watched the Olympics, partly because I simply had trouble watching any of the players on Team USA and actually rooting for them. There are those like Kobe who I simply don't like much, even if he's always a fun watch. There are those like LeBron who I still dislike heavily, despite not really hating him anymore. I love Tyson Chandler, and I love Andre Iguodala -- fantastic guys, fantastic players. But Paul is incorrigable and Williams is worse. Anthony Davis still causes me to raise an eyebrow. You go on down the list, point by point, until you reach the Thunder kids. I don't like Westbrook, and most of my friends don't like him much either, so that doesn't seem all that rare. I dislike Harden despite having respect for some aspects of his game. But, well...
It's nothing personal. It's kind of stupid, actually. I think Durant's likely the nicest, least-controversial star in the history of the league. He's got an "edge" only insofar as a plastic butter knife technically has one too. He's smart, engaging, and his personal story (streets of DC, putting OKC on his back, leading a team of upstarts to incredible heights with an atrocious coach for three years running) is phenomenal. But the way he eliminated my favorite team -- a flurry of insane clutch performances, crazy three point bombing, and (in general) an absolute domination few can exact on any team in the modern league? It was vicious. Eventually, I'll be able to watch him without thinking of that. Hasn't quite happened yet, though, even when watching normal regular season footage. Which leads to a general antipathy and a lack of any desire to really watch him play. You kind of know what you're going to get, in a sense.
Which leads to the main point of this post. A lot of people refer vaguely to the concept of youth when they discuss the Thunder, and default to the position that the Thunder will "naturally" be better next year due to youth alone. And thus, I posed a question to myself. How much better can Kevin Durant really get? I thought, and thought. I thought some more. And, after crunching the numbers and trying to figure out his most likely steps forward, I came to a basic conclusion. He could be a bit better. He could be a bit worse, if his shooting numbers fall off a tad -- something that often happens to tall shooters earlier than most. There are a few basic ways to improve his offense -- he needs to either get better at getting himself open against pressure (something he's currently horrible at) and better at controlling the ball when he handles it (as he's extremely turnover prone). Beyond that, short of becoming a Nash-style sniper, it's hard to see how Durant improves his offensive game. He can shoot from any spot on the court, he draws free throws in bundles akin to the Walter White methylene barrels, and he has a better shot than anyone his height in the history of the league, excepting (perhaps) the immortal Dirk. He was the best defensive rebounding small forward in the league and every single problem he has on defense is rooted in his general body-type, not a lack of effort.
But the most likely scenario? He stays about the same, onwards to the infinite.
• • •
When I say that I think Durant is roughly what he'll be in the end, I'm not saying there's no chance of the other possibilities. It's possible Durant's shooting has been overachieving a bit, and that it'll temper off. It's also possible Durant will get better -- I outlined a few ways above, like a better sense of how the hell a player should assert to get open in a pressured situation. It's a confidence interval, with a reasonably large confidence that Durant stays about the same and a small tail at either end for the other two options. The thing is, when you look at essentially every big piece on the Thunder, you start to see a similar picture. The Thunder are young in years, but old in experience -- everyone but Harden has played more than enough minutes in their career to consider their developmental period over (or at least highly close to it) and their peak years beginning. There are certain individual things each player could potentially work on, but in terms of wholescale revamping of their games, there aren't a ton of realistic possibilities.
Which in a general sense is why I feel the consensus view on the Thunder isn't quite accurate. It's not impossible, mind you -- I keep harping on uncertainty, but it's always worth repeating. It could happen. They could take the proverbial leap. It just isn't probabilistically likely, not at this juncture. The idea that the Thunder are simply bound to improve because they're so young is fundamentally flawed -- there are many ways to gauge a player's age, and when a player is operating at or near their prime performance and has played enough minutes, calendar age matters far, FAR less than their age in minutes. And there isn't a single member of the Thunder core that hasn't played more than enough minutes to enter their prime. Once in their prime, it's rare that you get a Steve Nash-style mid-career improvement. More likely, you'll simply see players perform at the same superstar level for a few years, gravitating around their actual mean performance with occasional months beyond or below that.
That's simply how aging tends to work in the NBA, and simply how the game is played. You rise, you peak, you fall. And when you reach your complete picture, you tend not to deviate too far from your productive peak mean. Which applies collectively, as well -- all this is essentially to say that I don't think the Thunder are going to be a markedly improved team this season. Eric Maynor should help, certainly, and Perry Jones is a fun X-Factor. But I don't think you can really count on the Thunder being improved. I think the Lakers will be improved, I think the Spurs will be about the same, and I think the Heat will be worse. But I really don't think the Thunder are going to be all that much better -- they'll be slightly different, perhaps, but not a whole lot better. Examining individual aging curves makes you wonder where the magical "they're young, so they'll improve" mindset comes from. These are all gut feelings, not statistical predictions. But, last season, I made sure to publish a pre-season thought regarding the Miami Heat and their ability to win a title, both going forward and in the span of last season. I'd like to repeat that general trend of getting my prediction down for future mockery, jeering, and (rarer still) the occasional right call.
So, yes. To offer my prediction as to how the Thunder's season will go, I think they start the postseason by getting pushed by a plucky eight seed that they nevertheless dispatch in 5 or 6 close games. I think they win an easier-than-expected series against the Grizzlies or the Clippers, and proceed to beat either the Spurs or the Lakers in the Western Conference Finals. They'll win in different, unpredictable ways. Perhaps their athleticism leads Brooks to run the Lakers into the ground. Perhaps age bites the Lakers too quickly, perhaps Nash has 2 or 3 off-games, perhaps Kobe goes iso-crazy at exactly the wrong time. As for the Spurs? Perhaps they win by holding homecourt and exploiting the permissive San Antonio defense. Who knows, really. It could be either, any, or all. But I think it's got a high probability of happening.
And then they'd meet the Heat, once more. Except this time, LeBron is a year older and Wade is a year worse. Their supporting cast -- full of stodgy old codgers and dodgy Juwan Howards -- is a year older, and their "big additions" from the previous offseason are worn down through the rough grind of an 82 game season and the simple machinations of Father Time. I think the Heat will be worse, next year, at least when the chips are on the table and the title defense reaches its pressurized peak. And throughout this all, will the Thunder be better? I don't think so, not to a significant degree.
But will they win a title? Quite a different question. And yes, Virginia -- I do believe they will.
• • •
I'm currently about halfway through Nate Silver's brilliant new book, The Signal and the Noise. I'm looking forward to finishing it, and I think I'll try to write up a decent-sized review/appreciation of the work. One aspect Silver focuses on in the early chapters is something I -- in college -- distinctly remember calling professors on with regularity. Many commit the sin of overinflating their own views, and making their ideas out to be certain beyond the reality. You'd think students of statistics would be mostly immune to this. After all, we learn confidence intervals and uncertainty principles at every step of our development. We spend hours and hours calculating these things by hand, and every single professor harps it until the day is done. Don't overstate your case. Don't go too far. Don't bloviate.
But alas, we're hardly immune. In fact, we may at times be worse off -- often, we simply think we're immune due to all the time we spent studying it. But it's no professional thing -- it's a human thing. Statisticians of all sorts find themselves prone to an unfortunate overconfident bluster similar to the type that any old TV analyst, social scientist, or politician espouses with impunity. We create a strong argument, a strong theory, a strong view. Then in our belief we buckle down and forget all those safeguards and caveats that any analyst should be actively sharing with their followers. As I did last year, when I -- again -- conflated confidence in the logic with certainty in execution. In the rubble of just how much that loss hurt, I never really had the chance to apologize for my overconfident reasoning leading some of the people who enjoy my analysis astray.
So, back to that wrong prediction.
My thought process was thus: the Spurs had (over the past 3 seasons) utterly dominated the Thunder in their individual matchups. The Spurs' defense -- regular season though it was -- had done a very effective job cutting down on fouls and keeping the Thunder off the line. The Spurs' offense, moreover, was essentially unguardable for the Thunder. No team in the league (not even the lowly Bobcats!) had allowed a higher offensive rating to the Spurs over the past several seasons. Going into the Western Conference Finals, the Thunder had won in San Antonio only once in the entire history of the Thunder franchise -- a meaningless game in late 2009, if you're wondering. This was the basis of my prediction -- without home court advantage against a team that had regularly obliterated them over the last 3 years, I simply didn't feel like the Thunder had a high probability of manifesting a close series. And the thing is? It wasn't a bad prediction in any of the individual steps of logic.
That all was sound. Everything fit. I seemed to have a good grasp on the pieces, a wide bredth of evidence to my assertions. So what was my problem? Why do I feel the need to apologize in any sense? Simpler than what you might think. What was bad -- and where I as a forecaster failed my reader -- was in the level of confidence I felt and openly assessed at this prediction. I told just about everyone who'd listen the litany of stats that proved my point, in turn becoming increasingly confident in my predictions and all the less uncertain in my assessment. In doing so, I managed to commit a cardinal sin in forecasting -- I conflated the strength of my evidence to the certainty of my prediction. You simply can't do that in the prediction business, because like it or not, you're going to end up off the mark sometimes, for reasons that simply don't fit prior expectations. The Thunder displayed more defensive acumen than they'd displayed at any previous point in their development in last year's conference finals, getting quality contributions from just about everyone on the face of the earth. There were games called so poorly I nearly had an anyerusim. There were shooting slumps, bad plays, and an uncharacteristic lack of confidence from the boys in black and silver. There were problems I simply didn't see coming. Ghosts in the machine, things that always seem to happen.
So yes. I don't apologize for getting it wrong -- I do think it's best practice to actively call out the times you miss the mark, if only to give added transparency to the people who follow your work, but it's not something to apologize for. It's something to call out. What I do apologize for is that tricky, annoying, and misleading overconfidence that led me to be remarkably wrong rather than simply slightly wrong. That's where I strayed, and that's where an apology is necessary and needed. And with all that said? While I don't think Durant has much of a leap left to make, I want to emphasize that he just as easily could. Analysis isn't prescription -- prediction isn't a death knell. Nothing final, nothing crazy. Never put an excess of trust in a single analyst.
And, well. Note to self: I certainly shouldn't do that either.
• • •