Integrating Technology: Ice Hockey (Part 1)
On April 23, 2015, a company called HockeyTech posted a press release about their new analytics system (You can find it at http://www.hockeytech.com/hockeytech-introduces-groundbreaking-analytics-system/). Through a local GPS system, HockeyTech claims they can track every event in a hockey game and provide statistical “ammunition” for teams to make decision on players – who to keep, who to cut and how much to pay them.
Of course, although this is all about hockey, there is great application to so many other sports: soccer, basketball, field hockey, rugby, even American football.
I am going to lay my cards on the table right now and say I am a little old school when it comes to how I approach coaching the game. I am skeptical of too much technology and heavily relying on statistics to make decisions – I think it can be a slippery slope. And, over the past two seasons, statistics have become a huge part of the coaching game in hockey. Most NHL teams have now hired one and sometimes two statistics “guys” to crunch numbers and analyze the game – and much of that data comes through software platforms like HockeyTech’s.
For the most part, coaches trust their eyes and what they see. As video coach for Canada’s national women’s hockey team, I provide a chance for our coaching staff to trust their eyes a second time when they review “game tape”. While taping the game live, my job is to code or mark the video with every event that takes place on the ice. On average there is an event that is marked every five or six seconds resulting in anywhere from 650 to 750 events per 60 minute game. There are a number of software packages that can provide this functionality. At Hockey Canada we use an excellent product called STEVA which has versions of their software for multiple sports. Ultimately, there is a mark placed on the video whenever the puck changes possession, when the puck goes from one zone to another and when there is an event such as a shot on net or a goal. There is much more functionality in the software that can be used when watching the game a second time (the game moves too fast the first time through to catch everything) including things like mapping where the shots on goal were taken from and where at the net they were shot.
Ultimately, video analysis is used for two things: evaluating your own players and scouting your opponent. This past season we added an iPad to our arsenal of tech tools and had a second coach charting line changes and shift length in real time. Not only was this a valuable statistic for coaches to determine minutes on ice for each player but the real time charting was fed into the video system so that coaches were able to isolate one player’s shifts if they desired. Post-game, the video feed and markings were all uploaded to the “cloud” so that players could access the game on the internet and watch their shifts back at the hotel or at home.
Scouting our opponents was also made much easier with the use of the system. Typically we would video tape and code our next opponent’s previous game. There are certain things that our pre-scout coaches will be looking for. For instance, the coach that is in charge of penalty killing will undoubtedly want to only watch our opponent’s power play clips. This can easily be sorted when the game has been coded in advance. As a video coach I find it very powerful to watch a number of clips of an opponent in succession. For example, being able to see 40 of our opponent’s forechecks in a row gives you much better insight into their systems and tendencies than picking them out while watching the entire game again. So in these two respects, player evaluation and scouting opponents, video analysis has become a crucial teaching and learning tool for players and coaches.
As mentioned, the other big trend in the hockey world is the use of what is called “advanced” statistics. These statistics go beyond the goals, assists and penalty minutes that we have typically collected. Now, hockey statisticians are collecting anything and everything, and crunching the numbers in every way, in order to quantify the game. Statistics unto themselves aren’t “technology” but the collection of those statistics can often be tedious if done by hand – much easier if collected automatically by sensors, etc.
One of the best applications of this technology is illustrated in this short two minute video produced by USAHockey (https://www.youtube.com/watch?v=CB_Ygapyl7c&feature=youtu.be). Working with players eight years old and younger, they studied the effects of making the playing surface smaller (I know, the soccer world figured this out decades ago!). Using the same type of technology that HockeyTech has now introduced, the results were that young players playing on an appropriately smaller sized playing surface had to perform many more critical skills in any given playing segment. For instance, there were twice as many puck battles in a smaller player area, six times more shots per player, twice as many puck touches per player and five times more puck receptions per player. These are astounding numbers and their collection was facilitated by technology applied to the game.
Again, a lot of the technology that is being used is directly related to collecting statistics and much of the use of statistics in hockey have been loosely modelled on the “Moneyball” model of baseball’s 2002 version of the Oakland A’s. That season, Oakland general manager used statistical models to put together what turned into a very successful season with very little money to spend on salaries. Over a decade later, the hockey world (or more correctly the NHL) wants to find ways to evaluate players and their relative salaries based on statistics. Ultimately, for professional hockey, this becomes a cost savings vehicle as much as anything. Now players will be evaluated more and more on their statistical “values” and less on the intangibles such as leadership, character and work ethic.
In part 2 of this multi-part article, we will take a look at how technology is helping players improve their game – from using iPad captured video at practice to reducing the risk of injury by monitoring load levels.