Penn was part of the management team that helped build the Grizzlies from a league doormat into a team that made the first three playoff appearances in franchise history, despite the franchise not getting anything from the No. 2 overall pick in 2003. That year, they were pipped for No. 1 overall by the Cleveland Cavaliers (in the LeBron James draft) and then had to confer their (only top-1-protected) pick to the Detroit Pistons because of a 1997 trade for Otis Thorpe, missing out on potential stars like Carmelo Anthony, Chris Bosh, and Dwyane Wade (or taking Darko Milicic, like Detroit did). While in Portland, Penn helped orchestrate the progression from the infamous “Jail Blazers” era, peaking with a fifty-four-win campaign in 2008–09.
As someone who remains very much plugged into the league (and also is co-owner and president of the new MLS soccer team launching in Los Angeles as early as 2017) but isn’t a part of a franchise, Penn spoke openly on the importance of information transfer and value inside an NBA front office.
“It really starts first with leadership’s appetite for that kind of information,” Penn said. “And when I say leadership, you’ve got the owner, the head of team operations—whatever title you’re going to call that—and then you got the head coach. And if any one of those, particularly the head coach, is a roadblock, or is not open to this, it’s useless. It really comes down to where the rubber meets the road and where the decision makers are willing to legitimately give this stuff weight.
“Every team over the last fifteen years—doesn’t matter whether they believe in it or not—they do this in order to cover their tail and to demonstrate that they are sophisticated. But a lot of that can be a ruse, or an empty effort if you truly don’t have the decision makers employing it. And I gotta tell you, it’s a lot easier said than done to employ this information and to figure out the right context to put it in because, at its essence, we’re dealing with human beings and performers who are more like Broadway actors or Hollywood actors than they are corporate management, or corporate executives, or corporate anything.
“And you’ve got the vagaries and the human side of it, the relationship side of it, and that always trumps 100 percent, no matter what. It just does. So the analytic components can be additive to a well-run human organization that deals with the relationships and deals with the frailties and everything that goes into this.”
Penn noted that the influx of many new technologies in the league was creating a different landscape from the one he was operating in just a handful of years ago. While much of the discussion about analytics involves the communication and sharing of information, the primary building block is the initial collection of the actual raw data, with teams having more equal access to the loads of information being generated.
Where once the advantage was in early adoption of the actual processes to collect data—like the Mavericks and others trying to stop Brian Kopp from bringing SportVU to every team—now the in-house advantages lie in how good the people are at parsing that data, and then how well and creatively you can break it down, communicate it, and then execute it based on your findings.
He also emphasized how fragile the entire process can be, with one misstep—whether it be in computation or interpersonal dealings—able to compromise large parts of the process.
“A few years ago, only Dallas, Houston, and Portland had data points; now the whole league has access to [them], but the cutting-edge folks are to the next phase of data collection,” Penn said. “You know, first it’s collection, then it’s analysis, and then it’s communication.
“So those are the three phases of it. It’s get the data, [and] figure out how to analyze it in a way that it’s helpful, but at every step of this, you have to have [a] communication aspect on the back end or it’s useless. And you can have your brilliant stats guy meet with your head coach once the wrong way and say the wrong thing based on some formula that’s silly, or all formulas have outliers; coaches are really good at drilling down and finding the outlier information. And then they will . . . a head coach who is busy and under pressure will just flat shut that dude off. He will be effectively useless to him, and that becomes useless data.”
Golden State Warriors assistant general manager Kirk Lacob concurs. Lacob, the son of team co-owner and venture capitalist Joe Lacob, has been in charge of ramping up the Warriors’ analytics efforts. On one hand, given how far sports use of increasing amounts of data has lagged behind other industries (which Lacob witnessed as his father built his fortune), he is fascinated by how much attention this area is getting in the basketball realm. On the other hand, he understands perhaps better than most that communication is the key to the value of the info.
“We’re learning a lot, and people are certainly analyzing the data that’s suddenly been dumped on us at a kind of an increasingly rapid rate,” Lacob said, “but . . . no matter what data you have and how well you put it together, if it’s not used properly, it just doesn’t happen.
“It really doesn’t matter at all. Because you can have all the data, but if it’s not being used for one reason or another—and, in this case, because people who need to use it don’t understand it or don’t want to see it—it’s ‘What are we doing the whole thing for?’ Because what we’re trying to do is narrow the margins on the edges, and, you know, slice off that last little piece of advantage we can find, and if it’s too much of a headache to slice that piece, then sometimes it’s better to just be more efficient and, you know, take your 5 percent error rate rather than cut that error rate to 2 percent.”
Lacob, whose father also was a blackjack player, also knows Jeff Ma, a member of the famed MIT blackjack team that was depicted both in the book Bringing Down the House and the movie 21. He uses analogs from that casino game to describe the NBA team-building process. The goal, per Lacob, is to keep refining your strategy as you learn more and more about the environment you’re in, and then be ready to “increase your bet” when the situation is favorable for you.
Of course, in basketball, you’re not just playing against a dealer (and the house). There are twenty-nine other teams in the same market for players or assets, so Lacob said, at times, team-building can be more like a blackjack tournament, where you are competing against other players with various strategies, but a lot of the final result comes down to who hits their big moves at the end. Having that kind of mindset is helpful because, as Lacob notes, a team can position itself extremely well, or get lucky and be in a seemingly advantageous position, but the cards—or, in this case, the draft class or the player performance—don’t end up rewarding you.
An example of this is the Cleveland Cavaliers, who, after having won the draft lottery in 2011 and nabbing blossoming star point guard Kyrie Irving, won it again in both 2013 and 2014. The 2013 pick, though, came in a draft where there was zero consensus as to who the best player was, or whether any of the players in the draft would be difference-makers as pros. The Cavaliers tabbed forward Anthony Bennett, who was dreadful in his rookie season in Cleveland. The following May, after beating the odds yet again, future star Andrew Wiggins was waiting for them at No. 1. The Cavaliers subsequently packaged both Bennett and Wiggins to the Minnesota Timberwolves for standout forward Kevin Love after luring LeBron James back as a free agent, so what was initially a good luck/bad timing outcome in 2013 became much more favorable after they repeated the trick with better timing.
Lacob also noted that the San Antonio Spurs have benefited over many seasons from having the same management and coaching personnel in place, as well as their top players, so the Spurs’ communication across the entire franchise is more refined than any other team in the league. Given Golden State had a new coaching staff in 2014–15, with rookie head coach Steve Kerr at the helm, figuring out the proper dynamics of communication were even more crucial, and this was something the franchise really focused on from the outset of the season as Kerr’s staff looked to build on the fifty-one-win success of prior head coach Mark Jackson. Every team approaches this in its own way, but wit
h the Warriors, the front office wants to empower the coaching staff to feature as the lead voices.
“To me, it’s important to talk to the people who are actually going to do the work,” Lacob said. “So what does that mean? It’s the players, the coaches—the coaches have a direct line to the players, [so] in the front office, we never want to get in the way of that. I mean, look, there’s, I’m sure, some teams [where] that information should be conveyed directly from the front office, maybe that’s how they believe it [should work]. We don’t. We think the coaches are the ones working every day, [so] we as management do not have any interest in talking to players every single day about this stuff.
“We want there to be one voice, because it’s the voice of nineteen on the court, when things matter. So the coaches are the ones delivering this message to the players. How can we best deliver this message to the coaches in a way they can understand? And, you know, we really pride ourselves on trying to find that balance, and it’s a very delicate balance. We’ve struggled with it at times, no doubt; we’ll be the first to admit that. But I think we found a pretty good spot, and now we have a new coaching staff, who we’re working really closely [with], and one of the things we kind of told them from day one was, look, let’s find out what you’re comfortable with, what each guy is comfortable with, and [though] we want to share what we want to give you, we also want to hear what you want to see, and let’s kind of try to find a happy balance between that.”
Lacob also concurred with what the Raptors’ Rucker said in terms of it being harder than you would imagine to discern exactly what analytics is doing for you on a night-to-night, or even month-to-month basis. When you’re dealing with small edges that you’re trying to make slightly larger over time (and sustainable), it can be difficult to isolate individual things happening, even as you believe the entirety of the operation is working to win you a possession here, or a quarter here, or steal one extra game there.
“A lot of times even [with] data, you look for, ‘Does there seem to be incremental improvement somewhere?’ And, you know, then we have to test out whether it’s randomness or not,” Lacob said. “But you’re really not going to see the full effects for probably—on most things—for a couple of years. And, you know, for game-to-game stuff, it might be over a course of a couple months. You do have to let randomness play out, and you do have to take that kind of variability and kind of test it out and see—how big are the variants here? But, no, sometimes we won’t really know, and so, we’re making a bet. Absolutely, we’re making a bet on our future.”
There may be no bigger bets in the NBA than player acquisition, and the most expensive ones come from the free-agent market. There, a player’s current team usually has an advantage in terms of being able to offer extra years and guaranteed salary, and all thirty teams are jousting in a controlled market for very limited amounts of talent. When you layer in individual team systems and coaching preferences and agent/franchise relationships and everything else that goes into whether one team has a chance with a choice free agent, you shouldn’t then be surprised to learn that while analytics has a very crucial role in targeting players, it can’t have nearly as much impact in terms of what price you end up paying for the player, regardless of what his “valuation” is.
As explained by multiple front-office people, player acquisition is a binary exercise: you either get a player, or you don’t. As such, while you may value a primary free agent as a “$10 million player,” the open market—and possible advantages to the current team in terms of contract size—often create market dynamics that drive a player’s cost up beyond his perceived value. At that point, teams need to evaluate whether that cost can fit into their salary structure, and how badly they want this specific player versus secondary options. A lot of the buzz every summer, and now especially with the salary cap rising so much and so quickly over the next couple of seasons, is that a guy is “overpriced” when he’s signed. For certain decisions, though, that’s a more favorable outcome than missing out on the player altogether.
“Very infrequently would you think about ‘let’s not give this player this much money because of some analytical nuance,’” Penn said. “The talent is so scarce, and the opportunity to get talent is so infrequent, that almost always you gotta go grab the talent when you can. So, the numbers would play into the overall feel of who’s the right talent and who’s the right target, but very rarely would you get into parsing out some detail, some point of the analytics, and say, ‘No, we’re only go to pay x, not x + whatever percentage.’ That’s more of a straight-up negotiation thing; ‘this is our guy, let’s get this done.’ That’s what we used to do. We’d pick our target, and the analytics would help us shape who our target is, but then we would spend as aggressively as we could within the rules to get the deal done.”
The related complication is that timing within a club’s development is a huge factor. There may be a free agent that you really like, but he doesn’t fit with your other key pieces. It’s not a rotisserie or daily fantasy team where you simply have to fit pieces in within a salary structure. You have a partially completed puzzle, and you’re constantly evaluating what pieces you need and what are available along with what the cost considerations may be. In the end, analytics may suggest a piece isn’t exactly perfect, but if a player comes available at the right time, and you can afford him, you may need to pull the trigger anyway.
“You can have all this theoretical analysis—we did this, we analyzed the last ten or twenty [teams], for the last ten years of finalists—and we looked at the way they were built with their cap structure,” Penn said. “Was it two great players? Was it three great players? Where did they invest in what positions? And what that ended up outputting was a really interesting study for a sports management class, but it’s completely impractical because you never really get a blank slate, and you never really get the opportunity to go out and build like that. You sort of, by circumstance, bump into or arrive at whatever talent you got, and then you have to build around whoever that is. ‘Best player available’ is a phrase used a lot.”
Of course, the players are the main actors that make your analytics decisions look good or bad. And while there are a good number of players in today’s NBA that take analytics output seriously, there’s still probably no better overall case than Shane Battier.
Battier, who retired after the 2013–14 season after a thirteen-year career, became the player face of the analytics movement after a lengthy Michael Lewis New York Times feature in 2009 unveiled a lot of the information that showed Battier to be a really valuable piece of winning teams even though his personal statistics were never particularly impressive.
“I was fortunate enough to play in an era that the analytics were able to explain and give . . . more validation than if I had played ten years earlier,” Battier said. “From college ball all the way to the analytics movement, I was always described as a glue guy. And the hustle guy—a guy who played hard. It’s a nice way of saying he’s an unathletic guy who helps the success of the team. Basically a proxy for ‘glue guy.’
“And, really, with the Michael Lewis article that came out about me, it was the first time that someone could explain—and Lewis did an amazing job—to explain what I did and why I was an effective basketball player, even though even most basketball pundits couldn’t put a finger on exactly what I did to positively impact the game. I was always part of winning teams and championship-winning teams, but my numbers were never overly sexy, my skill set obviously was not overly sexy, but I was always on the floor when it mattered, and more often than not, my teams won.”
As widely reported in recent years by the sports media, the Houston Rockets are involved early in a lot of the deeper analytical thinking in the sport. And in Battier’s case, Morey and the Rockets smelled out his hidden value before anyone else. In a 2007 Houston Press column on Morey’s methods, Jason Friedman wrote that when the Rockets’ brass were conducting a self-analysis prior to the 20
06 draft and seeing who they could acquire to fix some of the issues, Battier “stood out like a Mensa member at a Paris Hilton party.”
The Rockets went on to trade the No. 8 overall pick (and forward Stromile Swift) to the Memphis Grizzlies in exchange for Battier, which drew predictable howls, especially since UConn’s Rudy Gay had dropped in the draft and was available at that spot. It’s probable that even Battier didn’t understand why the Rockets had given up that kind of asset to acquire him, but that soon changed. So did the ways in which Battier understood the game, and how he could best deploy his own abilities to further his team’s chances of winning.
“I didn’t even know what the analytics really, really meant. It wasn’t until I got to Houston that I had it sort of explained what my value was to the Rockets’ organization, and it gave me a perspective on my career,” Battier said. “And even when I was playing before the Rockets in Memphis, I did things that I was taught to do—just be a good teammate, move the ball, block out, make the opponent take a tough shot, run back on defense—all the things that translate into modern-day analytics as the price of winning.”
“It really was an amazing experience to play for [Daryl] Morey and Sam Hinkie, who sort of explained their worldview of basketball, and I’ve always been super-analytic and always tried to look at a situation as it is versus what it’s supposed to be or getting caught up in the hyperbole. And the way they drew me into their analytic world—and I don’t know who the player was, it might’ve been Kobe, but they said, ‘Hey, do you know who Kobe Bryant is at his core?’
“And they showed me this scouting report on Kobe that he was a dominant right-handed player that finishes and gets fouled and shoots a free throw rate at a legendary level, and they had all the numbers and stats to back that up. And [they] really deconstructed one of the greatest players of all time, instead of traditional scouting reports that say Kobe’s got a really good right hand, he’s got a pretty good step-back jumper, he’s good in the post, all very general basketball terms, but nothing that could really give you an edge as a defender.
Chasing Perfection: A Behind-the-Scenes Look at the High-Stakes Game of Creating an NBA Champion Page 16
