The data revolution in cycling
Data is powerful, it can change the way people look at things, by bringing a deeper understanding of the matter. This is no different for cycling. Cyclist can better understand their mind and body by analzing the data they captured during and after a ride. Did you feel like your legs were heavy? Look at your data and their might be an aswer for why it was this way. Much of the data analysis is already done for us, or can be done for us, and is ready to be interpreted.
Using data to analyse how a rider is doing, is not a new idea. But the way fans, friends and family are experiencing the race, is changing rapidly these days. Those people are often the ones who keep the spirit up when the rider has some difficulties to go through, and support them along the training, resting and racing. And I think that’s a good reason why they deserve a VIP viewer experience. And what better way then to enrich the experience with the use of data?
And that’s exactly what was first happening at the Tour de France. But now you can see it emerging in mountainbiking too. Let’s take an example: the Absa Cape Epic 2018 in South Africa that was organised from the 18–25th of March. It is an eight day stage race, with over 658 km of racing, 13 530 m of climbing and four sequetive days of over 100 km of riding. For a summary of the race, please watch the video below.
But in which ways is data incorporated into the viewer and rider experience? According to Dimension data:
Friends and families can interact live with loved ones throughout the race. The mobile race hospital is connected to the Mediclinic national hospital network, enabling world class field triage and aftercare treatment. Medical histories are kept on record so that riders who compete again are given pertinent attention. Logistics vehicles and riders are tracked. Fans can see where their riders are on the route, where they finish, and what their ranking is. Riders can upload their own statistics to assess their performance on the go. Innovation is continuous. In 2015, we included video conferencing with British commentator, Rob Warner, at the course hot spots. In 2016, we expanded to live coverage from the route and on-screen data nuggets. In 2017, a Race Centre app was deployed, along with a PowerBI tool for real time analytics. Heart rate and power data for the professional riders was indicated on the race’s website.
But how do these things work?
First of all, there is a small device attached to the saddle of each rider, collecting GPS coordinates every second. The data are then streamed to a plane that is circulating above the race, using microwaves (which enables long distance transmission). These signals are subsequently sent back to the Data truck that is positioned at the finish line of the race. After the individual data reached the truck, data analysis can begin.
How the data analysis is done exactly is Dimension data’s secret. However, the riders location data is enriched with data about the weather, terrain, gradient,… If available, then also heart rate, power,… can be included, which provides even more information about the performance of a rider.
I can imagine that the data analysis that is done after collecting the different variables of a rider, included predictive modelling. This can help a team decide which rider is ready to race and which one is not. There is also an app that gathers information about health an well-being of a rider. In the morning, riders are asked a number of different questions, not only about their physical status, but also about their mental status. Afterwards, the result of the questions is combined with the data collected during training or racing. Looking at the data, the performance of a rider can be followed up closely.
The app is designed in such a way that it takes a minimal amount of time to answer the questions. The app is also used to gain scientific insights about the impact of travel across time zone and sleeping at altitude. Furthermore, the moment they went to bed and woke up and the quality of sleep is assessed. The app also records if riders train on a day, and if not, why.
But very interesting, there are also mood related questions in the questionnaire. Like how their mood is, the perceived effort of a training, and what’s their view on the previous training. And since the training data are available from powermeters, heart rate monitors, bike computers, … the real data, and intended effort can be compared to the rather subjective feeling and perceived effort.
Combining all information, the team hopes to develop a early warning system for illness and injury. By treating these things as early as possible, it can prevent riders from not being able to start in a race (or dropout from a race).
The dimension data team is trying to integrate the training data with the data collected in the app, and inject it back into TrainingPeaks, a software used to plan the riders’ training.
Now, a Dimension data quote that reveals a glimpse of the future.
Looking forward we want to apply machine learning and predictive analytics to optimise rider management. These insights will help us to keep all the riders performing to their maximum potential.
So, what does this have to do with what we at Arinti are doing, you might ask? Well, we have secretly been helping the guys at InTheRace build their app, a high-tech realtime performance monitoring system for cyclists.
With InTheRace, we believe we’re helping create a product superior to what’s already on the market. First of all, it’s a lot easier to set-up: there’s no need for helicopters, plane’s or trucks that create ‘moving networks’, we use a standard 4G connection to stream the data into the dashboard. That way, we can collect data during trainings as well and our data analysis will become better and better over time. Secondly, we believe our algorithms and models are more powerful and more accurate than what we’ve already seen.
So, what will the future bring for InTheRace? You might have read our previous blog series about stress monitoring. Think about how you could translate that into the cycling world. What if you could monitor stress levels of racers in high risk situations? What if we could prevent mass falls and injuries in the peloton? Keep an eye open for our future blogs.
If you liked reading this article, you might also want to read https://medium.com/sports-tech/datascout-about-using-data-in-sports-9442ca125730.
This blog was originally posted here.