Twelve Days, Five Sports, One Body: When a Laboratory Wears a Sports Jersey

Three in the morning in Nha Trang. I open a link. On the screen, a heart-rate trace runs across the Ionian Sea, and below it a GPS coordinate blinks at Ormenio, the northernmost tip of Greece. No scoreline. No standings. Nobody to beat. Just one man, one bicycle, one swimsuit, and a stack of sensors strapped to his body streaming data home in real time.

I am used to sitting in front of different numbers. In 2026, when the pandemic shut the pitches, I opened Excel — an entire Women's U-20 World Cup folded into twenty thousand rows, and I picked through every one. But what I saw on tonight's link did not resemble anything I had read before. People were calling it a feat. People were calling it a conquest. What I saw was a purely technical question: can a human being's physiological data be transmitted, stored, displayed and read correctly — while that person is swimming open water, climbing a slope, riding through rain, and the signal is patchy?

That is the part worth discussing. The rest is packaging.

THE ROUTE, AND THE MONEY BEHIND IT

The project is attached to a traverse: from Ormenio, the northernmost point of Greece, down to Gavdos, the southernmost island of Europe. Twelve days. Thirteen administrative regions. Five sports alternating — cycling, open-water swimming, mountaineering, running, sailing. A single fixed subject with no replacement plan: Giorgos Tsianos, a physician, a researcher, and an athlete. Funding comes from the Greek Ministry of Digital Governance and Artificial Intelligence, channelled through the Foundation of the Hellenic World, for an action titled "Integration of Artificial Intelligence in the field of Virtual and Augmented Reality, Phase B".

I stopped at that line. Because it told me what this project actually is.

It is presented as an ultra-endurance feat. But the budget line points somewhere else entirely: AI plus virtual reality, AI plus augmented reality, Phase B. That money belongs to a digital-technology budget, not a sports-science budget. The traverse plays the role of testbed, and also the role of stage.

I say this not to diminish anyone. I say it to put everything in its correct box.

Across seventeen years covering sport, I picked up a habit: whenever a project describes itself with adjectives, I go looking for verbs. "Feat" is an adjective. Transmit, store, display, read — those are verbs. And in the entire promotional text, exactly one sentence contains a real verb, a sentence the organisers wrote themselves and perhaps did not realise was the most important thing in it: whether data can be transmitted, stored, visualised and reliably interpreted in real time, despite the limitations of movement, weather, water, terrain and unstable connectivity.

That is the best sentence in the whole document. And it is an engineering question, not a slogan.

A LOAD-ACCUMULATION PROBLEM, NOT A RACE

Women's football does not need to be approved of as entertaining. It needs to be watched with a tactical eye. I keep that principle when I step into a field far from the pitch. A project like this also does not need to be approved of as epic. It needs to be read with a technical eye.

So what does a technical eye see?

It sees a load-accumulation problem pushed to its extreme. The fundamental difference between a twelve-day traverse and a marathon or a stage race lies here: the stressor is not a sprint, it is the rate of switching between movement modalities combined with fatigue stacked day over day.

Picture a body required to jump repeatedly between five different load profiles. Cycling is concentric loading, the knee turning thousands of times with little impact. Downhill running and mountaineering are eccentric loading, quadriceps and calves lengthened while contracting, the type of muscle damage that takes longest to recover from. Open-water swimming is thermal loading, the body simultaneously generating heat and fighting heat loss in cold water, with shoulders rotating continuously. Sailing is operational loading — heavy on organisation, light on metabolism, very likely serving as a recovery window.

The core of this project lies in the rate at which load profiles are swapped, not in the volume of any single profile.

That is why a 2:05 marathoner might not be able to do this, and a Himalaya climber might not either. What is being tested is not the ceiling of one sport, but the body's ability to recover when pushed into another sport before it has recovered from the previous one.

I once wrote about Japan at the World Cup in Russia, about how a sweeper dropped unusually deep to create a defensive mechanism nobody noticed because the scoreline hid it. Japan did not run faster. They used one player as a sixth chess piece, and the win came from the mechanism, not from the speed. The same applies here. What deserves analysis is the rate of load switching, a thing invisible on every results table.

THE BIGGEST PROBLEM: NOT A SINGLE NUMBER

But then I have to say this part, because it is the part that stops me signing my name to any performance assessment.

The promotional text offers no quantified figure at all beyond two things: the number of days and the number of regions. Total distance, no. Distance per sport, no. Cumulative elevation, no. Water temperature, no. Sea state, no. Target versus actual per day, no. Split pacing, no. Not one metric.

A physical effort without a single metric cannot be positioned on any coordinate system.

I am not saying this to nitpick. I am saying it because in my trade everyone has heard the line "if you can't measure it, you can't manage it". A project described as having high scientific value, yet publishing not one number about itself, leaves the reader to fill the gaps. And when readers fill gaps, they usually fill them in the most generous direction.

What I can say with certainty: given the structure of five sports, thirteen regions and twelve days, the land distance plus open-water swimming plus sailing almost certainly falls into the scale of a multi-stage ultra-endurance effort, meaning hundreds of kilometres on land plus significant sea swimming. But that is my inference, not the project's data.

This is the boundary I set for myself: when there are no numbers, I write that there are no numbers. I learned that from an older male colleague who once told me that women writing about football only needed to describe emotions, and that tactics were a man's business. I did not argue. I went home, rewatched the last ten matches of the Phong Phu Ha Nam women's club, and drew the high-press chart myself. I heard that sentence in a dressing room. Now I tell it back with numbers.

The only way to answer a prejudice is to be twice as precise. And precision begins by daring to say: this part, I do not know.

AI, VIRTUAL REALITY, AND THE REAL PRODUCT

Back to the budget line.

The funded action is titled "Integration of Artificial Intelligence in the field of Virtual and Augmented Reality, Phase B". The most important words in that sentence, to me, are the last four: virtual and augmented reality. Because they suggest the final product this money is aimed at is not a physiology paper but a visualisation product — a digital model, a visual platform, an immersive experience.

This changes how I read the whole project. A project designed to produce a visualisation product follows different logic from a project designed to produce physiological knowledge. It needs beautiful data, continuous data, representable data. And to some degree, it needs a compelling story more than it needs the full truth.

The words "Phase B" also deserve a pause. Naming a phase implies Phase A has been completed and a Phase C may exist behind it. That means funding is disbursed in tranches, and the next tranche depends on whether this one delivers. To an outside observer, this is a signal about incentives: the project has a motive to present success, and that motive does not necessarily align with the motive to publish failure.

The text itself calls this an ambitious project of high scientific and technological value that must be promoted and completed. The word "completed" sits in that sentence. In an arena where nobody eliminates anybody, where there are no quotas and no qualifiers, the success criteria are defined by the project itself.

When people grade their own exam, the reader must find a different marking scheme.

THE COST OF TURNING ONE PERSON INTO A LABORATORY

This is where I want to go deep, because it is rarely discussed in sports coverage.

The subject of the project is simultaneously its designer, its researcher and its public face. In research methodology, that is called an n=1 design with a subject-observer.

On the plus side, this design delivers exceptionally rich self-reported data and near-perfect compliance. You do not need to persuade anyone to eat on schedule, hydrate on time, wear the right device. The person does it because the person is the project.

On the minus side, this design has two unfillable holes.

The first hole is generalisability. One person does not create a trend. One person creates a story. You can describe one individual's fatigue curve in enormous depth, but you cannot say anything about humanity in general.

The second hole is more serious: interpretive conflict of interest. When the person collecting the data, interpreting the data, publishing the data and being measured are the same person, there is no mechanism preventing unfavourable data from being quietly set aside. This does not mean that person will cheat. It means the structure has no gatekeeper.

In the entire promotional text, there is no named research ethics board, no named independent medical monitor, no named scientific lead, no name at all beyond the subject. The team is mentioned twice with the phrases "great co-athletes" and "a broader network of qualified collaborators", but not one name appears.

For a project funded by a ministry and publicly disseminated, the absence of named scientific leadership is anomalous. In research, names are credibility. A paper with no author is a paper that does not exist.

I ONCE SPENT NINE MONTHS DOING SOMETHING SIMILAR

I understand why a single-subject design is attractive. I have done something close to it.

In 2026, when competitions stopped, I spent nine months rewatching 214 matches from the Women's U-20 World Cup between 2026 and 2026, carefully logging the changing role of the female full-back. My finding: female full-backs shifted from defensive sweeping to high advancing entries into the box, with run frequency rising 43 per cent after 2026. I compiled it into a forty-page report and sent it to an international women's football site.

I mention that not to compare. I mention it because I understand the value of sitting a long time with a dataset. But I also understand its limits. I had data from 214 matches, which means I had a large sample. With only one match, I would have had a good article and no conclusion at all.

The Greek project chose the opposite road: one subject, twelve days, and a great deal of ambition. I do not object to that choice. I only say it must be called by its correct name. It is a case study, not a physiology study. And a case study is very hard to defend before a review panel.

ONE PILLAR, ONE COLLAPSE

Within the project design, one thing reassures me more than anything else. And one thing worries me most.

The reassurance: physiological data is collected continuously — cardiovascular, respiratory, thermoregulatory, oxygenation, glycaemic, movement, work output and recovery. If this system works properly, it is a genuine safety net. It allows early detection of a body heading downhill before that body announces it by collapsing.

But here is the question I cannot answer: is that system used as a safety instrument or as a data-collection instrument? The two purposes do not exclude each other, but they lead to different behaviours. As a safety instrument, alarm thresholds are set low and someone has the authority to stop the attempt. As a data-collection instrument, thresholds are set at the highest tolerable level, because data at the limit is the valuable data.

The worry: the entire project stands on one pair of legs. No contingency plan is mentioned. If the subject tears an Achilles, inflames a plantar fascia, damages quadriceps on a descent, or suffers a systemic event such as hyponatraemia from over-drinking, hypothermia in open water, or heat illness on a land leg, then it is not only the journey that stops. The science stops, the broadcast stops, and the deliverable to the ministry stops.

A project with a single pillar has a single failure point.

Probabilistically, across twelve days of continuous multi-modal exposure, the chance of at least one significant physiological event is high. That is the nature of exposure, not a prediction about a specific individual.

HEALTH DATA BROADCAST LIVE — THE PART NOBODY MENTIONS

This is the section I most want readers to remember, because it rarely appears in sports writing.

The project will collect and publicly broadcast, in real time, the following data from a personally identifiable individual: cardiac function, respiratory function, thermoregulation, oxygen saturation, and glycaemic dynamics.

Under European Union data protection law, health data and biometric data belong to a special category requiring explicit consent and heightened safeguards. Live-streaming the heart rate and blood glucose of a named, visible, identifiable person is one of the most sensitive data disclosures a media-facing project can perform.

The promotional text describes the broadcast mechanism. It does not describe the consent mechanism, the anonymisation mechanism, or a retention period.

One detail changes the analysis: the subject is also the project lead. His right to anonymity is, in effect, self-waived. That reduces the severity of the consent issue. But it does not remove the requirement for a documented data governance framework, especially when the funder is a ministry specialising in digital governance.

Another layer: AI is named in the project as the means of scientific recording of biometric data. Under Europe's AI governance framework, AI systems processing health data may fall into higher-risk categories, carrying requirements for documentation, human oversight and data governance. That the funder is the Ministry of Digital Governance and Artificial Intelligence makes it plausible that governance documentation exists. But the promotional text does not supply it, and that silence is notable precisely because the funder is an AI ministry.

As for maritime safety on the swimming and sailing legs: entirely absent from the description. Open-water swimming and sailing across Greek waters, including a crossing to Gavdos, Europe's southernmost point, require coastguard coordination, permits, sea-state abort thresholds and rescue capability. These exist in any competent operation. They are simply not mentioned.

THE COUNTER-ARGUMENT: COMMERCIAL VALUE DOES NOT LIVE IN SPORT

At this point I have to turn in a different direction, the direction I believe is the project's real one.

The named applications include remote health monitoring, operational safety, research, human performance, and public understanding of physiology. In that list, remote health monitoring is the largest market, and it has almost nothing to do with elite sport.

In other words, the real commercial value of a remote measurement system that survives twelve days of extreme conditions does not lie in helping anyone run faster. It lies in proving that physiological signals can survive water, mountains, rain, vibration and patchy coverage.

And here I see a common confusion. People look at a project like this and think about medals. But its real competitors are not other athletes. Its real competitors are wearable brands and remote patient monitoring platforms.

I hold a fairly firm view about the transfer market and about intermediaries in sport: the noise they generate distorts the true value of the people working on the pitch. Here there is a similar kind of noise, except the noise comes from technology language. The word AI, the words augmented reality, the phrase high scientific value — these are budget-opening keywords. They are not evidence.

Evidence would be an open dataset, a methods paper, or a working product. None of the three appears in the promotional text.

If the measurement system survives twelve days and the results are published, this is valuable field evidence for the wearable industry. Multi-environment extremes are exactly where consumer devices usually fail: water, altitude, motion artefact, lost connectivity. A stack that holds up there is a stronger marketing statement than any brochure.

But if the results are withheld for commercial reasons, the project's public value evaporates, and what remains is a twelve-day advertisement.

ANOTHER RISK: NOBODY TO COMPARE AGAINST

One feature of this arena caught my attention. There is no ranking, no reigning champion, no quota, no qualifier. Which means there is nobody to overtake.

In competitive sport, an athlete's greatest risk is elimination. Here, the greatest risk is running out of money before delivering the goods. These are entirely different risk types, and they produce entirely different communication behaviours.

Elimination risk teaches humility, because someone else measures the outcome. Funding risk teaches presentation, because you define your own outcome.

That is why, reading a project like this, I always look for three things. First, the name of the scientific lead. Second, the method with instruments and sampling rates. Third, a commitment to open data. All three are absent.

I have sat long enough with data to know that a good dataset is not afraid of being looked at. What is afraid of being looked at is a dataset that has not been cleaned, or a dataset curated to tell a story.

THE ONE THING I BELIEVE

After everything I have just analysed, there is one thing I still believe.

The technical question the project sets itself is a real question. It is specific. It can be wrong. It can be proven impossible. And that is what makes it the best question in the entire document.

Can physiological data be transmitted, stored, visualised and reliably interpreted in real time, despite movement, weather, water, terrain and unstable connectivity?

If the answer is yes, it changes how we monitor people in places without doctors. It means a rig worker, a firefighter, an athlete on a remote island, or a patient in the highlands can all be monitored without someone sitting beside them.

If the answer is no, we still gain a valuable conclusion: where the limits of wearable devices in field conditions actually lie.

Both outcomes are useful. That is the mark of a good question.

And I am thinking about something else. In women's sport throughout my years in the trade, the hardest part was never a lack of talent. The hardest part was a lack of chroniclers patient enough for numbers to survive across seasons. A passage of play that is not recorded effectively never happened. A feat without a log is a spoken legend.

This is where two stories meet. The Greek project is trying to do what Vietnamese women's football still lacks: record systematically, so that later there is something to read back.

WHAT TO TAKE HOME

I do not live in Greece. I live in Nha Trang, and my work is women's football, women's basketball, women's volleyball, and the numbers nobody bothers to record. But this project leaves me three things I can apply at home immediately.

First, an effort with no metrics cannot be defended against a sceptic. To raise the profile of a sport, the first job is to build a recording system, not a media campaign.

Second, when a project defines its own success, the reader must find a different marking scheme. I will never assess a project only by the adjectives it attaches to itself.

Third, technology does not create value by itself. It creates value only when someone dares to publish the failures too. A project that shows only its successes is not science. It is advertising with charts.

CLOSING

Twelve days will pass. There will be legs cancelled by weather, legs where the modality has to change, days when blood glucose drops and heart rate will not return to the right zone. And if the organisers dare to broadcast those parts too, I will be the first to watch to the end.

If the trace on the screen stays a beautiful green for twelve straight days, I will ask myself: is that data measuring a human body, or measuring the audience's patience?

A good question does not need to be answered in twelve days. It needs to be answered by an open dataset, a few years later, when nobody remembers what that day's heart rate was.

That is the part worth waiting for.