Interview with Arne Jaspers – Performance Data Analyst, AZ Alkmaar

Could you start off by telling us a bit about your background?

I’m currently the performance data analyst for AZ Alkmaar who compete in the Eredivisie in the Netherlands. I’m also completing a PhD project in conjunction with the University of Leuven in Belgium and also in co-operation with TopSportsLab based in Amsterdam. I previously studied for a masters in both human movement sciences and physiotherapy, with work focusing on training and coaching in triathlon.

 

So what are your duties on a ‘typical’ day at the club?

Usually start around 9am to have a meeting with the medical staff and physical coach to look at the data from the previous days to make an overall assessment to provide advice about the training session for the day. Around 9:30am players will go into the gym to conduct injury prevention/stretching programme. The players will also provide subjective information around wellness and muscle soreness, from which I report to the physical coach prior to the training session (to see if adaptions need to be made). We then prepare for the main training session by putting the GPS and heart rate devices onto the players which we then monitor live using the Catapult system. Key part of this is to break down the session into each individual drill, such as 11v11 games, and my role is to provide feedback to the coaches around whether the players have achieved the pre-determined targets we set. Post session I will analyse the external and internal load data gathered from the Catapult system and compile reports based on our periodisation model.

 

What are the main challenges you face when working within applied sport?

The main aim of what we are trying to do at AZ Alkmaar is narrowing the confidence interval for good decision making. Speaking with other practitioners, everyone is trying to find this Holy Grail to best describe the relationship between training load and injuries/performance. My point of view is that we need to measure load objectively and connect with training outcomes throughout the season like medical attention, time loss injuries or fitness test results in order to gradually improve your decision making regarding training load prescription. Also listen to your or the coaches’ gut feeling about what happens on the pitch. It can inspire, give new ideas and reminds you of thinking about the connection with the game. In terms of coaches’ interest, we have had different coaches during my time at the club, and there requirements can vary. For example, one coach just wanted a traffic light type system to provide red flags when a player is at risk, whereas another coach wanted detailed information using the live system about players work rate. We try to inform the players to educate them, however the player’s responses can vary from wanted to know lots of information and others just have confidence in us to prepare them to play in matches.

 

When and where did you first find out about Catapult?

It’s quite funny actually! I was previously doing an internship in physiotherapy before I was approached by Robbert de Groot (ex-AZ Alkmaar physical coach, now FC Groningen) to help out using the Catapult system. After a 3 month period, they asked me to stay on at the club and I managed to get a grant to study a PhD whilst carrying on my work with AZ. So in a way, I got the job here at the club thanks to Catapult!

 

Glad to see we helped out along the way! So how long have you been using the Catapult system for? 

We have been using the Catapult system for 3 consecutive seasons now, which has helped us to build up a large database on information that helps us prepare players for matches by providing external and internal load data.

 

Why do you feel that you (or the club) choose Catapult?

Our general director and former coach went over to Australia to look for new ways to innovate and therefore visited some Australian Rules Football (AFL) clubs who were using Catapult. They liked what they saw and decided to buy the system based on these experiences. Personally, I’m really happy with the Catapult system, in particular the support we receive for things like broken units or questions around parameters and the software. The response time is very quick and I find it quite easy to use for things such as scientific research also.

 

How has Catapult influenced your role at AZ Alkmaar?

Prior to using Catapult, we didn’t really know the external load that the players were undertaking. It helps us to understand the load of each drill that the coaches prescribe, which can show how intensive they can be. Previously, we wouldn’t have been able to identify the loading of things such as small sided games, in which we use multiple parameters to quantify the load.

 

How do you use Catapult from an injury prevention/rehabilitation perspective?

To prevent injuries, we tend to look at the PlayerLoad parameter and also the high speed distance covered. We use the data to look at the acute and chronic training load to help identify when players may be at risk. We have identified thresholds for these parameters so we know based on previous experience (over both a 7 day and 28 period, as well as the balance between both) that we may have to modify the load and enhance recovery strategies. On the other side, we also look at the non-starters and look at the PlayerLoad values to try and identify when they are unloaded, due to a lack of match load. This information has also been enhanced by linking in with our physical testing, mostly using a submaximal Yo-Yo test. For the rehab sessions, we use the Catapult data to gradually build up the player looking again at the acute and chronic training load to ensure successful return to play.

 

You’ve already mentioned about the use of the PlayerLoad and high speed distance parameters. Are there any other important parameters that you use?

Yes we have also been looking at the PlayerLoad zones in more detail, in particular the higher weighted zones which we term ‘peak PlayerLoad’ to identify the intensive parts of the sessions. Due to the high sampling rate of the accelerometer, we can trust this data when identifying player movements and impacts. We also look at high acceleration and deceleration efforts. To get a general overview, we look at the total distance covered and then break this down into high speeds (15-20 km/h) and very-high and sprinting speeds (>20 km/h). During the drills, we tend to look at the m/min covered and also the PlayerLoad/min parameters. We have recently been investigating the inertial movement analysis (IMA) parameters during small sided games. So overall, we have our core parameters but we are always looking in an exploratory way at new measures and methods.

 

Without giving anything specific away around players, do you have an anecdotal story about how you’ve used the Catapult system to reveal something that you wouldn’t have been able to detect otherwise?

We had one coach who used a particular type of shooting drill on the day before the match in which the wide players (including full backs) who had to pass to the midfielders and then sprint a long distance along the pitch to receive the ball. They would do 10-15 repetitions which resulted in around 200-300m of high speed distance. We used this data to say to the coach that we need to change this drill, as it was not suitable the day before game. The coach didn’t realise the load would be so high as the drill generally looked quite easy, so the Catapult system helped to educate in this example.

 

Finally, how do you see the future of athlete tracking?

The next key part will be the use of GPS devices in competitive matches (since recent rule change by FIFA) as previously we have had to ‘guess’ what the match loads have been. I believe the accuracy of the devices will continue to improve and use of methods such as heart rate variability and other internal load measures will help to improve our prediction of performance and injury. Hopefully in the next 10 years we will have a better idea of how to prescribe the most relevant load for each individual player.

 

undefined