Despite a propensity to ankle and knee injuries, I played basketball from the time I was 9 years old through my time at university (intramurals only). To say I remain a huge fan of the game is a bit of an understatement. I get so excited during this time of the year that my friend’s work productivity goes down at least 20% and my friend ‘may’ struggle to stay below the “gotcha” live game streaming threshold at work.
This time of year when the NCAA (university) basketball tournaments begin is coined March Madness in the U.S. The term dates back to the 1930s. March Madness has been described as an ailment where ‘the thump of basketballs, squeak of sneakers and the roar of the crowd’ are almost impossible to contain. My heart rate increases when I see the intensity the college players have for the game and how well it is played. So today I want to take a look at some great examples of how basketball data has changed over the years and what we can learn from that.
I wanted to start with a great article on how basketball statistics makes data analytics possible and insight accessible to more people. Unlike other articles with statistical models or technical terms and math, this article has some great data visualizations for more people that don’t like or care to understand math, machine learning and the like in great detail. The insights we learn from basketball data include shooting percentages for star players, where shots were taken, and fun facts like the number of miles a player runs in an average game or the number of passes for a particular team. I love when data is ‘democratized’ or made available to anyone, not just to coaches or scouts.
In addition to great college basketball insights, the National Basketball Association (aka – the pros) recently released what us data geeks would consider a big data set since it includes information dating back to the first NBA season in 1946. Stats.nba.com allows mostly fans to look at basketball data with any level of depth and breadth. This is a great start in opening up the data but for developers and computer programmers to maximize the data, we’d prefer it in machine readable format rather than its current web format.
One of the more exciting aspects of open basketball data is how teams are using the information to improve their performance. Before there was data analysis, players and teams were judged with gut-instinct impressions from watching a game. According to an Atlantic article, the Houston Rockets decided not to shoot long-range two-point jump shots after looking at their data. The Rockets made it to the final playoff game against the Warriors despite having lots of player injuries during the season. Their success might not have only been attributed to how they decide which shots to take, but it’s an interesting start to how teams might incorporate data into their strategy.
I get excited when data turns into insight and then actions, especially if that means I get to hear more net swishing during March Madness.