Thursday, June 21, 2012

Missle Defense tech is finding the next LeBron

One of my recent posts talked about the advent of "Big Data" and how huge information stores would be driving business strategies of the future.  Before I lose you to boredom, let me shift into a quick Hollywood reference to give some mental adrenaline.

In 2011 a movie came out called MoneyBall and it starred Brad Pitt and Jonah Hill.  It was a really interesting story about how the hero and manager of the Oakland As, Billy Beane (Pitt), used statistics to take a really average group of players and made them one of the most successful teams in baseball.  The plot centered on how revolutionary and heretical it was for a person to use Big Data instead of emotion, or "gut instinct" to put together a team.  As you would expect, the hero had to overcome all of the skeptics and odds.  But in the end, the data won out and the team of losers had set the MLB record for most wins in a row and even made the playoffs.

The movie was truly a window into the birth of the Big Data concept.  Billy Beane was able to use record sets numbering the hundreds and several thousands to make decisions on players.  These decisions included how to specifically match different pitchers to certain batters, who played better at night versus day, and many other things such as the likelihood of how many runs a certain combination of players would (not could) produce during a game.

As cool as the movie and the use of data was, keep in mind that at most Beane was only working with several thousand data points at one time.  Today, 10 teams in the NBA are using record sets in the millions and that's per player, per game!  I have to give a huge shout out to Fast Company, who published this article:

http://www.fastcodesign.com/1670059/moneyball-20-how-missile-tracking-cameras-are-remaking-the-nba

Unlike the MoneyBall scenario that used yearly statistical sets, certain NBA teams are using cameras originally developed to track ballistic missiles (!) to follow each player, home and away, during every game played.  For example, Kevin Durant (plays for the Oklahoma Thunder) is being photographed and analyzed 25 times per second by a SportVU camera.  Rather than paraphrasing, let me quote the article directly:

"SportVU can tell you not just Kevin Durant’s shooting average, but his shooting average after dribbling one vs. two times, or his shooting average with a defender three feet away vs. five feet away. SportVU can actually consider both factors at once, plus take into account who passed him the ball, how many minutes he’d been on the court, and how many miles he’d run that game already."

Big data is not just some corporate phenomena.  In the NBA the data being collected is so specific and powerful that it is completely redefining not only which players are the best, but all of the conditions, movements, and heretofore unknowable attributes that make some people successful while others fail.  Think of how sports teams will pay more or less for players when they can analyze every single aspect of each person's performance on the court, every game for a full season. 

I would hardly be surprised if computers are not quickly integrated into each coach's game plan.  Imagine the Thunder's coach, Scott Brooks, saying something like this, "Well, we're playing the Spurs tonight and the computer says that they consistently beat us in rebounding.  I'll counter that by playing Kendrick Perkins more tonight and scale back on Serge Ibaka's minutes because, over 50,000 simulations we will likely pick up six more offensive rebounds, draw four more fouls, and shoot eight more free throws."  I can see a scenario like that happening as soon as next year.  Then, imagine where SportVU integrates with an iPad application used by the assistant coach on the bench.  The cameras will be taking into account injuries, fatigue, physical capabilities, refereeing, and a hundred different data points as they occur during a game.  Can you imagine the computer doing the analytics and making real-time game strategy decisions for the coaches?  I can.

The sad thing about the advent of big data into sports is that it will take some of the mysticism out of the games.  Want to know whether Jordan was better than Doctor J and don't want to include the number of championship rings?  Big data will give you the answer.  Want to know if the Celtics of 2008 could have beaten the 1984 Lakers in seven games - yep, there will be an answer for that.  Want to debate the intangibles of why Tim Tebow is seemingly such a bad quarterback but wins so many games?  Just check the data and you'll get an answer.  Big Data is the forbidden fruit of the new information era...

Given that the NFL is the biggest sporting franchise/organization in the world, you can expect that Big Data will be a big deal there soon, if not already.  Big data is just now starting to equal Big $Money$.  Given their propensity for process optimization and use of data/statistics, I think I'm going to go buy some stock in UPS...*


*Never, ever, ever, ever, ever take financial advice of any type from SimpleCIO!

No comments:

Post a Comment