Cullen: A Weekend at the Sloan Sports Analytics Conference

Scott Cullen
3/9/2011 12:14:13 PM
Decrease Text SizeIncrease Text Size
Text Size

As those who are brave enough to follow my Twitter feed know, I spent a couple of days in Boston last week, attending the MIT Sloan Sports Analytics Conference; a gathering of bright minds to discuss a wide-ranging number of topics in sports and the use of analytics to provide insights.

While this was the fifth annual event, it was my first time going. I've read in the past about's Bill Simmons' involvement in "Dorkapalooza" and, with hockey analytics having its own panel for the first time, it seemed like the time was right to check it out.

As expected with topics like "Skill and Chance in the NFL Draft" and "The Moral Hazard in Long-Term Guaranteed Contracts" (and, as promised by Simmons), the conference was a male-dominated affair.

There were 1500 tickets sold, up from 1,000 a year ago, and it didn't appear that any more than two or three dozen attendees were women, yet moments after I arrived for Friday's first presentation, a blonde, towering in tall black boots walked past me and I had an inkling that maybe sports nerds don't necessarily have to come with pocket protectors. More on appearances later.

The first panel was moderated by best-selling author Malcolm Gladwell and included Houston Rockets GM Daryl Morey, basketball announcer and former coach Jeff Van Gundy, Mark Verstegen, the CEO of Athlete's Performance, and New York Giants defensive end Justin Tuck. Primarily, the panel discussed an idea that Gladwell suggested in his book, Outliers, that 10,000 hours of practice were required to reach an elite level in a discipline, whether it be chess, musical composition or sports.

The result, then, was the question of whether or not an athlete could be enough of a "natural" in a sport to reach an elite level without putting in that kind of time.

Tuck shared some interesting anecdotes. First, was a story about his high school days, when Jamario Moon (the journeyman former Toronto Raptor now with the L.A. Clippers) was the unbelievable natural athlete who could jump out of the gym and do whatever he wanted as a high school basketball player.

At the same time, Gerald Wallace (now with the Portland Trail Blazers) was the kid who was, unlike Moon, always working on improving his game and the result has been a career that is much closer to those at top end of the NBA, including an appearance in last year's All-Star game, even if he wasn't as physically gifted as Moon.

Tuck himself, admitted that he grew up in such a small town that he didn't have quality football coaching until he was 16, a point at which he embraced it wholeheartedly and ultimately surpassed so many who had better preparation on his way to becoming one of the best defensive linemen in the NFL.

Of course, Tuck is 6-foot-4 and 265 pounds with rare athleticism, so let's not take this to be a statement about how simply anyone can practice enough to become an NFL player, but there are others with special physical attributes that haven't reached the heights of Tuck, so the differentiating point is that being coachable and willing to work at the game can be what separates the great from the good and the good from the marginal at the highest level.

To that end, both Morey and Van Gundy decided that Tracy McGrady was the most natural talent they had encountered, but may not have reached his ultimate peak because his physical gifts allowed him to dominate so easily that he didn't have to work on the finer points of his game, those that could have possibly allowed him to experience more success when he reached a level (say, the NBA playoffs), when the advantage of his individual skills wasn't enough to overwhelm the competition.

Perhaps it was easy to throw McGrady under the bus, since he's a rare talent who hasn't had playoff success, but Morey and Van Gundy certainly would have had an up-close and personal knowledge of McGrady's work habits.

Gladwell summed up his position on the matter, saying, "A lot of what we call talent is the desire to practice."

When you consider that a great hockey player like Sidney Crosby may not possess the greatest pure physical gifts, but is so diligent and committed to working at his craft (remember when he was a face-off liability as a rookie?), that there is something to the value that attitude and motivation can play in determining a player's ultimate potential.

I would have liked to hear their take on Josh Hamilton, who seems like he has to have rare natural talent to have abused his body the way he did with drugs and still be one of the premier hitters in baseball, but it just didn't make it into the discussion.

A few other interesting points to come out of this panel:

Van Gundy, who generated buzz with his strong opinions and a sense of humour, said that Stromile Swift may have been the biggest waste of potential greatness that he encountered as a coach, but said it was eventually understandable because Swift just wasn't that interested in basketball.

Like any athletic kid who was on the way to becoming 6-foot-10, Swift was steered towards basketball and obviously experienced enough success to get drafted second overall (by Vancouver) in 2000, but didn't love the game enough to work at it.

Van Gundy was very matter-of-fact about it, noting that Swift made a lot of money, and has done well with real estate investments too, but basketball just wasn't his primary focus.

Van Gundy effectively took a starring role in this panel, telling the crowd that player could be "one of soft, selfish or stupid, but not two." Moments later, someone asked about Bonzi Wells, to which Van Gundy continued, "...or fat," eliciting laughter from the audience.

The determination from many on the panel, then, was that they would rather have the player they knew was coachable and would put in the work to be great, but acknowledged how difficult it is to determine that level of motivation in a prospect. It's easy to be motivated as a collegiate athlete (or junior hockey player) but who knows what a player will become once they have $20-million guaranteed to them?

Tuck commented that some players (like JaMarcus Russell) will see a $50-million first contract and think they are set for life, while others would see a $50-million contract and think of earning a potential $100-million for their second contract.

As an aside, Tuck also said that if highly-touted Auburn quarterback Cam Newton tries to play in the NFL the way he plays in college, with all that running, "He will get hurt."

Van Gundy, while appreciating players who were well-rounded, didn't necessarily want them on his team. Ideally, he would have guys "who played the game and chased girls," though he wasn't even that into them chasing girls, but it seemed more a reference to keeping their life goals simple. It still got a good reaction from the crowd.

When comparing a pair of players he coached on the Knicks, Chris Dudley and Charles Oakley, Van Gundy pointed out it was the guy who went to Virginia Union (Oakley), not Yale (Dudley) who was the genius on the basketball court. Continuing on his theme of single-mindedness (perhaps simple-mindedness?), Van Gundy said he always preferred a player that turned to the sports page first ahead of business or world news. There's a value to a player that knows his role.

Morey asserted that as players get older, their defence gets better, as long as they care. Eventually, physical skills will erode, but once a player establishes their place in the league, the game should slow down enough that they can improve on the defensive side, where hard work and smarts will tend to yield results.

From there, I spent some time in both the Injury and Performance Analytics and talking with Brian Macdonald, an Assistant Math Professor at West Point, who was devising an adjusted plus-minus system for hockey. More on this when I tackle hockey and analytics later this week.

My few takeaways:

Concussions were obviously front and centre for the injury discussion, on a panel that included injury expert Will Carroll and Chris Nowinski, the former WWE wrestler who has been very committed to research on former athletes' brains for the Center for the Study of Traumatic Encephalopathy at Boston University.


Not surprisingly, those who are interested in injuries feel like teams and leagues aren't as forthcoming as they could or should be. Caroll's point on the matter was that the information will eventually get out anyway, so there's not much advantage to be gained by hiding it for the hours, days or weeks that it takes to find out what's really going on.

As it was with some of the panels, there wasn't really an inclusion of hockey when it came to this issue, though that might have been for the best. With NHL teams' tendency to be vague, at best, and, at worst, dishonest about injuries, it would figure to only have frustrated panelists further.

An idea that Carroll has been touting for some time, involving biomechanics studies of pitchers' throwing motions, seems like a no-brainer when one considers how much money baseball teams lose to the disabled list every year due to injured pitchers.

After that, was a lunchtime discussion with Apolo Anton Ohno, hosted by ESPN's Ric Bucher.

While Ohno had interesting stories to share about training and competing at an Olympic level, perhaps the most compelling was an approach that saw him trying many little things (ie. change in diet, improved skate sharpening) in an effort to improve by the tiniest of increments, with the goal of accumulating enough tiny increments to give him a significant advantage.

In many ways, the use of analytics is going through a similar process, trying to help teams that are willing to use them to find an edge, if even by incremental amounts.

It's not as though a team is going to go from being worst in the league to best simply because they start using advanced stastistical analysis, but maybe it's the difference between making and missing the playoffs for a team on the bubble, or it's enough to get homefield advantage in the postseason. For a championship contender, maybe it's the fine line that makes the difference in winning it all.

After lunch, it was a look at Sports Gambling, the Source of Sports Innovation? and then The Decision: How Players and Teams Will Choose in the Future

The gambling panel was moderated by's Chad Millman and included fomer MIT student Jeff Ma, perhaps best noted for his role on the MIT Blackjack team (made famous in the book Bringing Down the House and the movie 21), and New York Knicks Director of Player Personnel Mark Warkentein.

Warkentein had a different insight into gambling, as one of his early jobs was as an assistant coach was with Jerry Tarkanian's UNLV Runnin' Rebels, so he spent a lot of time in the Vegas casinos, but it provided real context for the changes that have happened in sports gambling.

Succinctly, Warkentien noted that in the old days, most sports bettors looked like he does (middle aged, overweight, glasses) and now they look a lot more like Ma, young and professional in appearance.

The consensus of this panel was that for anyone interested in handicapping games, the right way to go about it was to generate power rankings based on statistics, thereby removing emotion from the calculation. As the Power Rankings guy, they might as well have been preaching to the choir.

Ma also noted that anyone who says that they are picking 80% or 90% winners in sports betting is lying. He made a point that 60% is very successful and that still means a lot of losses along the way.

Now that he's an NBA Executive, Warkentein knows the value of advanced stats, but realizes that not everyone, like coaches for instance, will be so open to new ideas, which means part of getting a team to use analytics involves selling the value of these ideas. His suggestion? "Instead of taking 26 math classes, take 25 and one drama class."

The comparison Warkentein made was to the O.J. Simpson trial, when having DNA evidence wasn't enough for prosecutor Marcia Clark because, despite having the evidence she needed to tell the story, she couldn't sell it to the jury. Effectively, statheads have to be able to convince the coaches and executives why the information matters and how it is going to make their team better. If it will make the team better, there's obviously more reason to listen to new ideas.

From there I joined in on The Decision: How Athletes Will Choose in the Future, a panel moderated by ESPN's Michael Wilbon that included Spurs GM R.C. Buford, agent Mark Bartelstein, former NBA Player Donny Marshall and Maple Leafs GM Brian Burke.

As hockey fans know, Burke is prone to making bold statements. Working at, I've seen enough of his press conferences to know the drill, but the Sloan crowd simply couldn't get enough.

As those from other sports, specifically basketball, couldn't fathom the thought of violence and intimidation as a strategy, Burke relished it, talking about things like the "clear demarcation between our top six forwards and bottom six" and "placing a premium on hostility."

While other sports questioned the fraternization of athletes with their opponents, Burke took the opportunity to point out how Carolina's Eric Staal leveled younger brother Marc Staal, of the New York Rangers, with a hit that kept him out of the lineup for three games, adding that Rangers GM Glen Sather said the part that people didn't necessarily see was Marc coming back onto the ice to challenge Eric to a fight.

It can be funny gauging the reaction of a non-hockey crowd to the kind of mayhem that is, at times, accepted as the normal course of NHL business and don't think that Burke doesn't play to those differences perfectly.

Talking about the media market in Toronto, Burke told the assembled crowd, "Even a fourth-line grunt is a star in our town," and talked about how much more media coverage there is of the Maple Leafs, even compared to the Bruins, who play in a good hockey town.

Possibly Burke's finest moment of this panel came when he said he likes having a coach that is tough to play for, emphasizing his point with the assertion, "The word 'fun' doesn't appear in the standard player's contract."

Since basketball has embraced advanced statistical analysis so much recently, I stuck around for Basketball Analytics, a panel moderated by ESPN's Marc Stein, with stats guru John Hollinger, former GM Kevin Pritchard, Celtics Assistant Executive Director of Basketball Operations Mike Zarren and Dallas Mavericks owner Mark Cuban.

Naturally, Cuban was a hit with the crowd, before saying a word, as he was wearing a t-shirt with the slogan, "Talk Nerdy To Me."

Once things got rolling, this panel confirmed that, while NBA teams use analytics, it's still a wide open discipline because teams don't all value certain skills equally. Cuban suggested that he's happy for as many stats as possible to be widely available, thereby allowing teams to analyze the data that exists, rather than each team trying to create their own data. That seems like a long way off.

Agreeing with Burke's oft-stated mantra (to those in the Toronto market, at least), Cuban doesn't understand why teams wait until the trade deadline to make deals. If it's a good deal, it's a good deal two weeks before the deadline too.

What Cuban also pointed out was that the NBA isn't an efficient market for trades, that not all teams to every other team equally, leaving deals to happen that leave other teams wishing they could have been in on the action. On another panel, this might have been a great time to ask about the Joe Thornton trade.

Cuban also told an amusing story about the first deal he tried to make as owner of the Mavericks. It was a minor deal and he mentioned it to a friend playing pickup hoops. The friend passed it onto his friend at a local paper and when the story hit the paper, the other team in the deal backed out, ticked off with the new guy in the league not knowing to keep quiet.

Panelists agreed that rumours tend not to scuttle deals, in fact once a name is out there, it generates more calls from other teams, checking the availability/price to acquire a player. They were talking basketball, but surely this would apply to other sports as well.

Maybe an underrated part of the panel, Celtics stats ace Mike Zarren made one of my favourite points about the player evaluation process, stating that scouting information can be used as part of analytics too.

In Saturday's Baseball Analytics, Tom Tippett, the Red Sox's Director of Baseball Information Services, pointed out that using data helps provide complete coverage in a player's evaluation. "You can look at the full body of's not scouting vs. data."

That may have been the overriding message to anyone promoting the analytics side of things. The human element of observation is never going to be eliminated from the process, so the way to get analytics more prominently involved in the process is to find the best way to augment the already-in-place methods of evaluation (ie. scouting).

The final panel on Friday was one of the most entertaining, Referee Analytics, moderated by Bill Simmons, and including Cuban, Sports Illustrated's Jon Wertheim and NFL referee Mike Carey.

Cuban was on his best behaviour, having been warned by the NBA that this constituted a public forum, making him liable for fines if he criticized officials, so he got out his frustrations by grilling Carey about how NFL officials don't call penalties in all circumstances.

Carey, who excelled in the face of criticism (such is the life of a referee), acknowledged that NFL officials won't call a holding play if they deem it to be immaterial to the outcome of the play, which made Cuban apoplectic -- he wants a foul, er penalty, in the first minute to be a penalty in the last minute, no matter what.

Carey was also the official during Super Bowl XLII, when David Tyree made the miracle helmet catch against the New England Patriots. Despite Simmons' best efforts to get Carey to admit he missed holding calls, or an in-the-grasp sack on that play, Carey stood firm, acknowledging (jokingly) that the one call in his career he'd like to have back was agreeing to be on this panel.

There were more panels Saturday morning, starting with Baseball Analytics. Naturally, baseball is as advanced as any sport when it comes to statistical analysis and talk of things like Pitch fX and Fielding fX seem like the wave of the future, leaps and bounds ahead of where other sports are when it comes to using stats to analyze what is happening on the field.

Part of the baseball panel, Jonah Keri just wrote a book called The Extra 2% about how the Tampa Bay Rays used Wall Street Strategies, always trying to find undervalued parts of the game that would allow the Rays to not only compete, but thrive in the American League East.

The Hockey Analytics Panel was next, but considering we're pushing 3,500 words at this point, that along with other hockey-centric tidbits will come in my next blog.

One of the more entertaining panels was made so by the addition of Burke. New Owners: Challenges and Opportunities moderated by Bill Simmons and including Boston Celtics owner Wyc Grousbeck, San Diego Padres owner Jeff Moorad and Golden State Warriors owner Joe Lacob.

Burke continued his tour-de-force, chastising Simmons for a softball question to Grousbeck, saying, "I never thought I would see you suck up the way you are today."

When it came to dealing with owners, Burke said that in all three of his GM interviews (he got all three jobs), the first thing he's told prospective employers is, "There are two hands on the steering wheel and both are mine." I've heard this from Burke before, but many in the room had not, so it got a big reaction.

Not long after, in fact, Simmons asked Grousbeck if he had ever touched (Celtics GM) Danny Ainge's steering wheel, to which Grousbeck, without missing a beat, replied (referring to the recent suspension of Brandon Davies), "If I had, it would have gotten me kicked off the BYU basketball team."

As Burke drilled home his frustration, in the past, with new owners that figured by December they knew who should be playing on the power play, he seemed to impress Grousbeck, who asked if Burke knew anything about basketball.

Grousbeck delivered his own zinger to the Massachusetts students in the audience. When discussing his purchase of the Celtics, Grousbeck quickly figured out that if the asking price was, in fact, the rumoured $360-million (in 2002) then, based on the team's cash flow, he would have about $12-million of leeway annually, either to re-invest in the team or to provide a cushion if some years weren't so good.

It seemed like a simple calculation and Grousbeck noted that, "Because I went to Stanford Business School, I made the deal. If I had gone to Harvard or MIT, I'd still be analyzing it." Burn!

Also, while it seemed strange to have Burke on a panel full of owners, it provided insight into the relationship between owners and the ones calling the shots for the team.

Additionally, Burke made sure to point out that he works for the richest ownership group in North American sports, as the Ontario Teachers' Pension Fund has more money at its disposal than any of the very wealthy individuals on the panel. That seemed to catch the audience off guard. That cute little hockey league does, at least in this instance, have some serious financial muscle involved.

The final panel of the conference was a discussion about the live sports experience and watching on HDTV, moderated by ESPN's Ric Bucher with panelists Bill Simmons, Mark Cuban, Jonathan Kraft (President, the Kraft Group) and Stephen Jones (CEO, Director of Player Personnel, Dallas Cowboys).

Of course, Jones could talk about all the bells and whistles involved in the new Cowboys stadium and Cuban came across as the brilliant entrepreneurial sort that one would expect.

A few highlights:

Cuban telling Simmons that, "While others are trying to live in the fast lane, (Simmons) lives in the past lane." No doubt, Simmons tends to wax nostalgic for the old days of watching the Celtics at the Boston Garden.

Kraft had an interesting idea about having all players mic'ed so that, eventually, the Pats would be able to provide that audio as an app for those at the game to access on their smart phones. Cuban was skeptical, citing the immense technological challenges involved, but I've long said that the best hockey-watching experience would be the X-rated version, letting fans know just how nasty it can get on the ice. Dare to dream, right?

There is obviously a challenge, given the cost, comfort and amenities of home to lure fans out to game, but Cuban made an interesting point. "If you think about the first game you went to, you don't remember the remember what made it special and that's the emotion and people you shared it with."

One additional point of Cuban's that might have been my favourite, is that Facebook is hardly the wave of he future. If it's somewhere that 500 million people already meet, it's the present, not the future. He also said that Twitter isn't the future and made a simple point: five years ago, Twitter wasn't a big deal, so five years from now, whatever is going to be that massive wave of the future is likely to be something that isn't widely accepted now.

Hearing Cuban's passion for his product, when it comes to the Mavericks, made me think that, if there is anything to the rumours of his interest in the Dallas Stars, the NHL would be lucky to have him. He's a billionaire who wants to win and is a bright and energetic guy; how many fans would like someone like that writing the cheques for their team?

In any case, coming away from this weekend was incredibly inspiring. Smart -- I mean, really, really smart -- people with great ideas about how to make teams better and, in some cases, how to make sports as a whole better is a great way to spend a couple of days.

From the development of analytics perspective, I go back to Zarren, who commented on the growth of the conference, from a point at which they brought their own stools to the front of a classroom at MIT to being on a stage at the Boston Convention Centre in front of more than 1,000 people and perhaps the size of the audience helps illustrate the progress of analytics in sports.

Zarren commented that not only the size, but the physical appearance of the crowd (generally looking rather business-like), showed progress too, noting that there were "a lot of pocket protectors" in their all-male audience that first year.

If there is increased acceptance of analytics in sports, it's for the better. Of course, as a stats nerd, I take that position, but I'm completely on board with the personnel people at the conference that recognize analytics play a part in the total evaluation of players and teams.

It's a matter of doing the best job possible integrating the analysis so that the information doesn't get lost on its way to the field of play.

(Hard to believe after all this, I know, but I'll have additional comments later this week, on the hockey presentations and analytics panel, specifically.)

Scott Cullen can be reached at and followed on Twitter at For more, check out TSN Fantasy on Facebook.

Cabbie on

New York-bred hip hop artist Action Bronson discusses his friendship with Kevin Love, his jumpshot, Mike Napoli's beard, obscure sports references and Derek Jeter's Brand Jordan commercial. More...

He has speed in his DNA, learn more about Olympic champion Donovan Bailey's nephew, Jaden and his success on the gridiron in the latest Powerade 24. More...

© 2017
All rights reserved.
Bell Media Television