Last week I was invited to sunny San Francisco to speak at the Sports Analytics Innovation Summit by Innovation Enterprise. Data analytics was the common theme and so it was interesting to listen to speakers from the entire breadth of sports clubs and organisation and compare discussions from the business side to the on pitch analysis. Data clearly impacts upon all parts of the business and there were common themes throughout the presentations regardless of the application, such as the constant of change and the importance of communication.
It seems the business analytics may be a step ahead of the performance side in their organisation and integration of data but there are of course innate differences between the environments. From a performance perspective we know the faces behind the numbers, the squad of 25 or so human beings that we are dealing with face to face, rather than the millions of faceless data points other departments may deal with.
At times the off field business analytics was mesmerising! There was the fan engagement example from the Sacramento Kings of providing an individualised VIP experience on game-day from number plate recognition at the car park, which then alerts a member of staff to greet the fan at the gate, to automatically discounting their favourite burger at the food booth they always go to at games. Their aim is for fans to only need their mobile phone at a game, no tickets, no cash, just the swipe of a smartphone to provide them this individualised experience all driven by CRM (Customer Relationship Management system).
The San Francisco teams were of course central to the conference and discussed their data analytics successes at length across a number of presentations, which included the San Francisco Giants’ innovation of dynamic pricing ticketing (changing ticket prices based on analytics) and the 49ers upturn in traffic, audience and retention with their website revamp (presented by the central NFL Club Digital Strategy analysis) and their #FortyViners success on social media.
Michael Clutterbuck of the Milwaukee Bucks presented ‘Building a Sustainable Analytics Department’ that included five key pillars that I think are relevant to all data analytics regardless of the content of the data:
He also included the following quote from a Coach for us all to consider:
“I’m really inclined to think the most important thing stat geeks can do in 2014 is not to develop new statistics, but do a better job of explaining the metrics that already exist.”
In general in the UK we tend to develop as the ‘Jack of All Trades’ Sport Scientist, who are responsible for managing the data collection, analysis, statistics and reporting as well as the practical elements, whereas my impression of the American model was more independent. In many cases they seem to recruit the practical based Athletic Trainers/Strength and Conditioners as well as the (self-titled) ‘stats geeks’ to manage the data. Of course this is a fairly sweeping statement and there are exceptions on both sides of the Atlantic but this was my general impression.
For example, there I was with employees of LA Lakers, who with Masters and PhDs in statistics and computer science expected to go onto be Professors and by their own admission never in their wildest dreams imagined working for one of the biggest basketball franchises. And yet hand-picked and tested out with some trial research projects, here they are now presenting to coaches their findings on performances, draft picks etc. Another renowned example of this model that springs to mind is the Seattle Sounders with ex-Microsoft analyst Ravi Ramineni joining Dave Tenney’s Sport Science team to oversee the data management and analysis. There data scientists certainly know their ‘python’ from their ‘r’, and I wonder if/how long it might be before we see more staff with pure data analytics backgrounds in English football.
Inspired by the success of Moneyball in baseball, it is an interesting time for basketball with the SportVU system (http://www.sporttechie.com/2013/10/31/sportvu-pulls-nba-analytics-forward/) providing tracking data that comprises X and Y coordinates for players and X, Y and Z coordinates for the ball (which as I understand is also made publically available, which seems too good to be true for bloggers and the amateur ‘stat geeks’). Philip Maymin presented a fast paced, mind blowing discussion of the potential of Optical Analytics in the NBA. Not just understanding the basic statistics but unlocking every potential situation such as how often a player converts a three point shot from a certain area of the court, following a certain pattern of play, when a teammate is open in a certain area and during a certain period of the game. Not of course just stats for stats sake, but relating these numbers more to game situation, tactics and the real questions of the Coach. He also emphasised that ‘big data’ is not just anything that does not fit into Excel and questioned if we are really dealing with big data in sport. On the other hand, speakers from the US Olympic Committee presented how their collection of big data and technology is shaping decision making in medicine and performance with their athletes.
Now it goes without saying that basketball and baseball are more relevant and engaging for the impact of statistics than football – we are dealing with more players, physical contact, bigger playing area, ball movement in possession etc. More similar to American Football, Brian Hampton the Director of Football Administration and Analytics from the San Francisco 49ers, discussed ‘What Actually Benefits Teams’. He spoke of when one of his coaches was pressed in a media conference for specifics on stats used by the team to determine pass/offence/defence success, he repeatedly stated ‘Winning’ and ‘the stat we use is wins’. This was the bottom line, do not get caught up with the ‘mindless stats’, what ultimately matters is winning and does a specific play or action improve your chance of winning. This idea of blending the analytics and coaching was also the premise of my presentation; ‘The Art and Science of Data Analytics in Football’.
Finally there was much focus on concussions, specifically with NFL given the large incidence in the sport. Many wearable technologies and researchers are looking into the diagnosis and monitoring of concussion, such as with an accelerometer in the helmet. Given it is a hot topic in English football at the moment too it will be interesting to see how these solutions develop overtime. The company MC10 presented their ethos for wearing technology of having electronics that conform to us and not the other way round.
Many thanks to Innovation Enterprise for inviting me to speak at the Summit.