Stan Van Gundy was a bracing presence at this year’s MIT Sloan Sports Analytics Conference. Neither a naysayer like ex-Toronto Maple Leafs GM and human Pluggers comic Brian Burke nor a fanatic booster intent on pushing process over results, Van Gundy offered something more than a counter-perspective to the advanced stats revolution: a humanist angle. Speaking at the Basketball Analytics panel, he reminded the audience that a coach is more than a conduit for an approach, whether analytical or not. He began by expanding on the way he treated 2-for-1 situations at the end of quarters while with the Heat and Magic:
I know that’s something that, by the numbers, we should do and we didn’t do it. And you can argue with this, but there’s another side to this that you have to at least consider. Whether you agree with 2-for-1 or not, [when] a guy races the ball up the floor and jacks up some horseshit shot, look at the other four guys on the floor. They don’t run back as hard. And now when you start the fourth quarter and a guy just did that in the third quarter, now the next guy gets the ball and says, “I’m jacking up the next one.” One of the things in coaching is you’re trying to create a style of play and a culture that this is how we play the game. Every time you make an exception to that and say, “This is how we play the game but not in this case: you’re allowed to throw up whatever crap you want,” then you’re breaking down your system a little bit.
It was an effective way to make the point that just because a coach has a set of data, and even a way to use that data effectively on the court, that coach is leading a team composed of individual players, all with their own approaches and ideas and prejudices and misunderstandings. Getting an entire team to buy into a philosophy takes more than strong support by data. It takes more than being right.
Coaches are, after all, professionals who have made their reputations by successfully getting diverse groups of people to buy into their idea about how to play the game, even when many of those people see it differently. You can’t tell me, for example, that the same approach to explaining analytics—or anything—is going to work equally well with Andrei Kirilenko and J.J. Barea. In a way, truth doesn’t enter into it. It’s about belief.
But things got a little strawman-ish when he veered deeper into the territory of working with a team:
There are coaches that are stuck on their system and there are people who are stuck on their way of doing things like, “It’s all gotta fit my analytics” more than they are on winning games. And the mark of a coach is not understanding the analytics. That may help you. As a coach, you’ve got to be able to go out and get a team to perform, OK? I know how important it is as a coach—and probably there are several coaches that do—that we limit layup attempts, that we limit free throw attempts, that we limit 3-point attempts. But some coaches can get their guys to actually do that and others can sit here like you guys and play it like a video game, but can’t actually get people to perform. The goal is to win, OK? We’re not playing video games here.
He was actually circling back to an earlier point he made about the audience (“A lot of you analytics people think that the game is a video game and so players will always react as your models say they will react”). I find his use of video games as an example both fascinating and a little off.
He’s clearly using “video games” as shorthand for something simplified, basic, dumb—a thing that makes you feel superpowered when in fact you’re just some drooling teenager on a beanbag chair. People, he’s saying, aren’t pixels or polygons.
I have no idea exactly what kind of experience Van Gundy has with video games—whether he’s talking about Double Dribble or Blazers vs. Bulls or NBA 2K13 (he’s not talking about NBA 2K13)—but there’s something to it, if not quite what he intends.
There are plenty of players who are great in video games, yet somehow less than great in the real NBA: Michael Beasley, Jamal Crawford, Anthony Randolph (he once won MVP of the Finals in a simmed Association I ran), Nick Young, and the list goes on. What they all have in common is that when you, the player, get to make decisions for them, their considerable physical skills seem to magically fall into place. For example, Beasley stops eating up isos on the left wing with endless head fakes and jab steps before taking fadeaway jumpers. Instead, he’s balancing his jumpshots with drives to the hoop and using his length to become a shutdown defender.
On the flip side, video games can’t quite seem to figure out what to make of a player like Andrei Kirilenko because so much of what makes him great comes from his creativity and trickiness off the ball. The artificial intelligence—even recent games like NBA 2K13—isn’t sophisticated enough to make Kirilenko take advantage of cuts along the baseline unless the play is being called for him. Simply put, making players in these games act human is a huge challenge. In some cases, it makes your job as the player easier. In others, tougher.
So Van Gundy’s right: video games can make it seem like players are just sets of ratings instead of living, breathing, often problematic human beings. But it’s important to realize that the shortcomings of sports video games are a failing, not a feature. With every passing year, the simulation aspect of video game basketball gets better and better. For a taste of that, look at this video explaining the way NBA 2K13 deals with the Horns set, layering options on top of options in an effort to replicate the kinds of read and react plays that are a hallmark of offenses like Rick Adelman’s corner and the triangle. Yes: you can just ramp the difficulty down and go nuts with Jamal Crawford if you so choose, but more and more, video games are striving to involve us in the complexity of the sport, not dumb it down.
It’s not quite the same in non-sports video games, though.
As the video game industry has gone more and more mainstream, there’s been a trend towards trimming back the unyielding difficulty of early games. Back before games could effectively tell a story, the primary appeal of them was ever-escalating difficulty. They were challenges to be overcome. But as the medium grew more concerned with story, with player experience, there’s been a rise in tutorials, in hand-holding, in ways to manipulate the game world without moral judgment.
You used to have to cheat to get 30 lives in Contra (say it with me) and then you were a cheater. But these days, the very idea of “lives” has been jettisoned from nearly every game. Designers want you to experience the whole game and studios invest millions of dollars in games that gamers want value from; the surest path to these results is to make sure gamers stick with the game all the way through.
In that sense, Van Gundy is dead on. As video games have increasingly focused on immersion, on experience, they have downplayed difficulty and consequences. Except for Dark Souls.
At first glance, Dark Souls might seem like a typical hack-and-slash fantasy adventure, but you don’t have to get further than the box to get the sense that there’s something different about it. In great big letters, it tells you: PREPARE TO DIE. And oh God will you die. Having just recently started it, I expected a certain amount of difficulty up front that would fall away as I gained the requisite experience and power.
Nope. Your character moves clumsily, especially if you weigh him or her down with the heaviest armor you can find, which won’t even be all that heavy to begin with. The swords take forever to swing; the bows are weak and ineffectual, good only for drawing the attention and ire of undead soldiers and skeletons. And if you don’t face down each one of those enemies like they’re a serious threat, they’re going to kill you. Repeatedly. The screen doesn’t simply fade to black, but nor are your myriad deaths overly cinematic. They are, however, unfailingly accompanied by deep red text rising on the screen and proclaiming YOU DIED.
And every time you die, the enemies come back. Hell, every time you even rest at a bonfire—the game’s shorthand for a save point—they come back. There’s not even any pretension to a full or thorough explanation of how this is possible within the game’s world. It’s just how it is, and you have to deal with it.
So what you learn from Dark Souls isn’t that someone has spent lots of money on this lavish game world for you to enjoy, has carefully crafted a story to make you feel heroic. It teaches you that all there is is the grind. Over at The Classical, Yago Colas penned a tremendous ode to the repetition that goes into making a shooter like Ray Allen great. That willingness to do things again and again is the writ-large version of the microwork of Dark Souls.
If you rush things, if you try to go through shortcuts, if your attention slips and you don’t treat every challenge with the gravity it demands, Dark Souls is going to kill you again and again and again. It can be immensely frustrating if you’re used to having the experience spoonfed to you, but if you adjust, you begin to find pleasure in the simplest things: you learn where your enemies lurk, you value the easy fights, like that dumb skeleton with the crossbow at the top of the tower with terrible aim. You learn that the dying is not something to avoid, but to embrace. You learn from a good death.
I know, I know. It’s kind of silly to liken the years of dedication and work that players and coaches in the NBA put into their careers to a frustrating video game, but I can’t help thinking that Stan Van Gundy might like Dark Souls. I think he could appreciate the way it refuses to concede to the prevailing trend in game design, but rather defines its own culture, creates its own style of play, and says, “This is how we play the game.”