GPS has become more and more mainstream in professional sports. The combination of improvement in technology, as well as reduction of cost over time, has made these systems more accessible to most pro and college teams. I think using GPS is team sport is important because training is too hectic and there is no other way to quantify volumes and intensities of work done. Team sport training is not nice and neat like track & field and weightlifting. Understanding the training load of “Sprint 6x30m” or “Clean 8×1 @ 80%” is far easier than understanding the amount of work done in 4x5min 6v6 small sided game for each of the 12 individuals. This chaotic training environment does make GPS a bit more valuable.
While valuable, also oftentimes misconstrued and used inappropriately. The biggest gripe I have is when performance coaches / sport science professionals put training ceilings on players because they believe they can’t handle the “load”. Many times this will come back to haunt teams because in an attempt to not do “too much”, the athlete probably hasn’t done enough. Not every non-contact injury is caused by doing too much. In fact, I would say many injuries occur because the athlete has not prepared enough to cope with the demands placed on them (i.e. training too little or not intense enough).
Using GPS volumes to predict if an athlete has done too much in an attempt to avoid injury is black magic. There are far too many factors that lead to an injury to boil it down to volume alone. Unless an athlete has consistently been injured when hitting a certain volume threshold, we can’t say that completing x amount of ‘insert metric’ in a week is any more injurious than completing y amount ‘insert same metric’ in a week assuming the athlete is generally fit and healthy. From a preparation standpoint, I do see a need for relatively consistent loading of volume to avoid detraining of general muscular-tendon capacity. As Martin Bucheit has written about, supplementation of HSR volume to avoid large fluctuations in training volumes may be appropriate when you consider an athlete might spend weeks out of the starting lineup and be inserted back in at any moment. But, to cut athlete’s off from training or to say this person is at a higher risk than someone else because they’ve done “too much” is a disservice to the athlete. The truth is, we don’t know what the athlete’s capacity is and rather than guessing what that ceiling is, why don’t we train with the idea to raise physical capacities and prepare the athlete for the worst-case-scenario if and when it does happen.
One way we can use GPS a bit more to our advantage from a performance-oriented mindset is to understand what these highest intensity running or worst-case-scenario moments of the match are in regards to a physical standpoint. Much of the current research on techniques to quantify and understand the highest intensity moments of the game up to this point has been done by Jace Delaney & Co. and seems like the most logical way to approach and understand this idea.
Using the current research, we can understand peak running demands by using a moving average calculation. The moving average calculation would measure the athlete’s distance covered over x minutes (you could also use HSR/min or Acceleration/min to understand more specific movement demands of running intensities as well as changes in speed).
The moving average calculation is more appropriate and robust than using set blocks of time because the moving average ensures the peak activity is captured whereas set blocks of time may miss the true peak intensity period.
|Blocks of Time|
You can clearly see that using blocks of time isn’t as useful is because peak demands of play might actually occur from time bin 3-8, but the separated blocks of time will not catch this, and this method usually underestimates actual peak demands.
Another way to see this would be a real-life example. I’ve plotted out a 1-minute moving average across the first and second halves. I’ve also circled in red the highest intensity running period for each individual (notice that it doesn’t occur at the same time for each person). If I had used the block time period method, we may have very well missed these actual peaks in activity.
After calculating peak demands for individual players, we can then subcategorize peak demands if we wanted to. I like categorizing by position to understand the absolute positional demand of the way we play. We could go further and separate by formation played, but I have yet to find any major differences in positional demands when accounting for formation. Even if we found some differences, I don’t think we’d have enough certainty to say increases or decreases in peak game demands are truly due to formation because there are many other factors (opponent formation, score, weather, surface, travel, etc.) that affect individual game demands regardless of formation played.
As much of the research has done, we calculate a 1 – 10 minute moving average because most of our drills fall within that time duration, but you could calculate and time period if you wanted.
If running demands are categorized by position, you’ll finish with something like this (only two positions represented):
Once we know the peak running demands, how do we use this to assist with training? We have used this information a couple of ways. On our preplanned intense days, one way is to attempt to set up drills that allow us to maintain or exceed peak match demands. Things to consider in doing this is the size of the pitch, touch restrictions, playing from the keeper or playing throw-ins / corners, etc. I think the biggest thing that I have seen influence training is that the players must understand that you want them to play intensely and to not put so many conditions on the game that the players are constrained from playing their normal style of play (i.e. how they would play in a match).
We then look at the training drill intensities and compare to match intensities for the same duration (i.e. a 5-minute possession game compared to the match peak 5-minute moving average). We can use this information live or retrospectively. As an example, we might see team running intensities decline over multiple sets, indicating that if we might need more rest between sets (assuming our goal is to train at or above match intensities). Or we might see individual running intensities decline or remain under the positional intensity threshold over the course of the given number of sets which might tell us that individual needs more fitness to maintain the level of match intensity for that position (or they need train with intent).
We’ve created the above to help relay this information a little more clearly. A simple, yet effective way to visualize running intensities for drills. With this layout, it allows us to compare multiple athletes within the same training session as well as individual athletes to their specific positional demands during intense match play.
Though everything needs context, and with this quantitative feedback we need to watch training to understand why there might be a decrease in physical output. If we’re playing 11v11 and the ball is consistently in Team A’s half, we might see Team B’s Center Back activity decrease significantly because they’re not involved in much of the play. It’s not because the CB’s are unfit or tired, but rather the opportunity to run is not present. Having an understanding of this type of context with the quantitative information is crucial to providing appropriate feedback to players and coaches.
So how should GPS be used? For performance or for athlete monitoring purposes? Well, it can and should be used for both, I just believe GPS has some major limitations when we make conclusions on an athlete’s state of risk based on volume of work done alone.
The pendulum seems to always shift from performance mindset to medical mindset and back to a performance mindset every so often. My question is why is the pendulum swaying at all? If you do the performance piece right, the medical piece (injury reduction) takes care of itself. The real problem occurs when the two are viewed as separate entities when they are, in reality, one and the same.