Maarten Enthoven already possesses physical numbers that most athletes could only dream of.
The Dutch athlete has completed a 2K row in 5:59 and averaged 440 watts for 20 minutes on the Concept2 BikeErg. He has a 54-minute HYROX Pro PB and competed in both Elite 15 Singles and Doubles races.
However, Maarten knows that to close the gap to the best HYROX athletes in the world, he needs to find an edge. His solution is “Project David”, an attempt to combine his experience as a data scientist with artificial intelligence to improve his training, recovery, nutrition and race-day decisions.
Running Training
Despite his huge engine, Maarten openly describes himself as one of the weakest runners in the Elite 15.
His 1:15 half-marathon PB is impressive by normal standards, but he regularly competes against athletes with much faster running times. Improving that weakness has therefore become one of his main priorities.
He has started working with a running club and coach, increased the amount of track work in his training and focused more closely on his technique.
One of his preferred sessions consists of alternating 200 metres above threshold with 800 metres at approximately marathon pace. He completes eight kilometres continuously, without recovery between the faster and slower sections.
The important part is learning to settle back into a strong pace after the harder effort. Maarten believes that resembles the experience of trying to run efficiently after a demanding HYROX station.
Use Data to Make Decisions
Maarten collects information from his heart-rate strap, watch and HRV measurements, while also recording his rate of perceived exertion.
However, he does not believe athletes should track something simply because a device makes it possible.
“The goal of data collection isn’t just to collect data. It’s to help you make a decision. If the data isn’t helping you make a decision, then it’s not useful.”
Before adopting a new test or metric, Maarten asks what it will change. Will it affect the session he performs, the intensity he selects or how he approaches his recovery? If the answer is unclear, the information may offer little practical value.
He also identifies three common problems with data-driven training: measurement errors, missing or insufficient information, and poor labelling.
A heart-rate measurement becomes less useful if the sensor was worn incorrectly. A workout file can also be misleading if you fail to record an unexpected break or identify what the session was intended to achieve. Good data requires context. Recording how a session felt, what you were trying to accomplish and anything unusual that happened can make the numbers far more useful later.
Turn Race Results into Race Strategy
Project David extends beyond training. Maarten believes race data can help athletes make better tactical decisions.
He points to the sleds at major championships, where certain lanes can move faster than others. Published timing data shows when athletes enter and leave a station, making it possible to identify lanes where competitors repeatedly lose positions.
Maarten experienced the consequences himself. At one race, he entered the sleds near the front but left around 12th after selecting a slower lane. Another athlete had watched earlier races and deliberately avoided it.
Maarten wants to analyse results from earlier waves so he can make a more informed lane choice.
The principle also applies outside of just looking at sleds. Reviewing your splits can show where you consistently lose time, whether starting faster improves your result and which stations damage the following run most severely.
Sled Pull and Lunge Speed
Maarten’s advantage is already visible on the HYROX stations. In one Berlin race, he recorded the third-fastest sled pull and lunges among 945 athletes.
He believes the sled pull is one of the most technical parts of HYROX. Athletes often waste time managing loose rope or stepping backwards before they begin producing useful force.
His advice is to create a direct connection between your feet, body, hands and the sled. Removing the slack immediately allows the force you produce against the floor to transfer into moving the load.
His lunges are another strength. While some athletes build volume with bodyweight lunges, Maarten takes the opposite approach and regularly trains with loads above the 30kg men’s Pro race weight. This conditions his legs to the load, although he accepts that the run afterwards can suffer.
Finding Limits
Maarten competes hard because he wants to discover where his limits are. That requires accepting occasions when a race or test does not produce the result he wanted.
“If you never fail, you never really find out where your limits are.”
He demonstrated that willingness during his rowing career, when he broke six minutes for 2km with a 5:59. He also uses off-season challenges, including ultra-distance runs and bike rides of more than 300 kilometres, to give his training a different purpose away from HYROX.
These challenges carry more fatigue and risk than he would accept close to an important race. During the off-season, however, they keep him motivated and allow him to test himself in a different environment.
Project David follows the same philosophy. It is an experiment rather than a guarantee. Maarten is trying to find out whether better information and smarter decisions can help him compete with athletes who currently possess superior running ability.
His approach provides a useful model for any HYROX athlete: identify the problem, collect relevant information, test a possible solution and use the result to decide what you do next.
You can check out the full episode on the Rox Lyfe podcast, or watch below…
