Interactive 3D Analysis

Benedikt, June 2 2023

When doing turns, technique changes or "going over the top" in a clever way, a cross-country skier can easily win a few seconds without increased effort. To better identify these key sections and find out which "strategy" is the fastest, we created the interactive 3D analysis. This is the first feature we developed in close collaboration with our partner Team Aker Daehlie. More will follow at the end of the summer.

Screenshot of the 3D profile with positions of each pole plant indicated by a colored dot

Saving Precious Seconds

Cross-country skiing is a very complex sport: the best, or fastest, race strategy is highly dependent on snow conditions, weather, and the athlete herself or himself. What will work well for one athlete in one situation might not work at all if the situation and/or athlete changes. Therefore, it is crucial to find out what is fastest in conditions as similar to the race as possible. For example, while training on the race-track the days before the race.

This requires a very fast and intuitive feedback, along with the possibility to explore different strategies. There are two ways to achieve this (I recommend combining both to get the most out of it).

Strategy A: Identify key sections apriori

During initial scouting of the race-track, based on your experience as a coach, identify the three most important key sections. For each key section write down up to three different strategies for maximizing performance.

Then let the athlete ski the race-track and for each lap provide him/her different instructions. Don't forget to have one lap where the athlete can choose his/her own preferred strategy. Make sure to minimize the effect of fatigue on the outcome...

Strategy B: Let the data speak

In a training race, let athletes compete against each other and/or do an interval training. You may provide a set of general instructions for each lap but otherwise don't interfere.

Find the Seconds

First, use the smart segmentation to automatically detect all laps and mark them for the in-depth processing.

Screenshot of the smart segmentation with automatically detected tracks
The smart segmentation automatically detects all reference tracks and marks them with different colors. The user can then mark the segments he/she wants to keep for in-depth analysis.

Second, get an overview of what happened and where the athletes lost / gained time with the race analysis feature. Do you see some unexpected time gains or losses? Where are the sections with biggest time differences between laps and athletes? For strategy B, identify the key sections you wish to analyze further, with the interactive 3D analysis.

Time waterfall plot providing an overview of the timing for the entire track.
This time waterfall plots compares each lap to a self-selected reference lap. The curves are "continuous" split times: instead of having a time information only 1-2 per lap our solution provides a split time for every two meters. Curves which go down mean that in this part the athlete was faster than the selected reference. Curves which go up mean that in this part the athlete was slower than the selected reference.
In this example we can see an impressive time gain of over 7 seconds during the second part of the large uphill for the light blue line compared to the reference. This could be a key section: over only 50 - 70m a potential time gain of several seconds can be reached (of course under the assumption that the athlete is able to deliver so much power).

Third, open the interactive 3D analysis via the dashboard. Mark the first key section you would like to analyze. Then select up to three athletes / laps you would like to compare against each other and click on "Analyze".

Screenshot showing the selected track segment for the interactive analysis
The start of the interactive analysis showing a satellite image of the track along with the elevation profile and a 3D profile. Here, the user selects a part of the track with the lasso tool. The selected part will then be analyzed in-depth.

For each selected lap the system creates an interactive 3D map with the elevation profile, color-coded sub-technique and marks speed differences in red or green. The exact location of each pole push is indicated, allowing to precisely see where and how much an athlete pushed. Moreover, four graphs show time differences (waterfall plot), and compare cycle duration, cycle distance, and speed. Interactively, you can now find out what condition was the fastest and how many seconds can potentially be gained.

Screenshot showing the 3d interactive analysis. The 3d profiles can be rotated and zoomed with the mouse.
The final results, with the three 3D profiles and additional graphs for cycle information. These graphs will then allow to understand what happened: how much time was lost or gained in this selected section? How large was the speed difference? Did the speed difference come from a different cycle duration (cadence)? Or was the cycle length (distance covered for each pole push) different?

This is great, but how long does it take to get these results?

Results can be obtained within a few minutes and you don’t need to be an expert in data analysis. Our algorithms automatically compute everything for you. If you are using the real-time feature, first results can be obtained immediately after each lap, during the training.

And, does it help? Yes!

"In our team we had one athlete who always missed sprint finals by a few seconds. By using this analysis, we discovered that precious time was 'lost' during turns. We could provide the right feedback and without increasing power output the athlete qualified for the next finals."
- says Dr. Hans Kristian Stadheim from Team Aker Daehlie.


Contact us for a free demo and to buy your own system!

And yes, these two guys were also very happy they could prepare their world championships with the help of our products. :-)

Two pro cross-country skiers holding our sensors