Southeast Asian Athletics After Aichi-Nagoya 2026: The Craft of Reading Empty Cells

In late October 2026, three windows were open at once on my desk in Shanghai. The first held a recording of an international athletics meet that had just closed in Japan, where a twenty-two-year-old hit the last hurdle of the 400 metres, fell, and sat on the track for a long time before an official came to lift him up. The second held the live data feed of a European football match: more than twelve thousand data points for a single game, every run, every torso angle, every breathing cycle interpolated and labelled.

The third was a file an editor sent me at six in the evening, with one short line attached: "Please read and add what's missing." The file had nine sections, forty-two table cells, and almost every cell carried the same phrase: N/A. Insufficient information to assess.

I sat with that file for a while. Outside, October rain fell thin, and the tram crossing the bridge hummed at a steady pitch. I wondered how an analytical sheet about athletics, a sport measured to the hundredth of a second, could be so empty. Then I remembered a morning at a stadium a few years ago, standing in the mixed zone past the finish line, watching a female athlete's calves still trembling in small pulses a full minute after she had stopped moving. The organisers that day published no wind reading. No splits. No reaction times.

On paper, I had enough data to file a story. In reality, I had half a story.

That is why I am writing this. Not to complain about an empty spreadsheet, but to describe a sport that runs on empty cells, and to ask what is being poured into them instead.

The framework and the silence it leaves behind

The template my editors sent is a standardised nine-part structure: event and performance; athlete condition; competition structure and qualification; event landscape and national strength; rules and anti-doping; team and training systems; risk landscape; public narrative and expectation gaps; and transmission into the wider sports industry. Each part has tables, cells, and a one-to-five-star rating system.

I know this kind of framework well. I once built similar models, only for football. In 2026, when I left a statistics firm to join a new sports media platform in Shanghai, I built a pressing model that measured Barcelona's pressing success rate by pitch zone. When Neymar moved to Paris for 222 million euros that August, I published proprietary data: Barcelona's pressing success rate fell from 34.2 percent to 28.7 percent over the first half of the 2026-18 season. A few coaches consulted that piece. I learned that data can tell stories, but only when someone is willing to release it.

Athletics differs from football at exactly this point. Football has enormous broadcast rights revenue feeding a data ecosystem: thirty-two cameras, optical tracking, third-party statistics firms, data contracts with licensed betting markets in some jurisdictions, and a fan base large enough to make selling data profitable. Athletics has no such structure. A regional athletics meet in Southeast Asia typically has one local timing company, one photo-finish system, one printed results sheet, and one PDF.

That PDF usually contains four columns: name, country, mark, place. Four columns for a race that needs at least twelve before I can read it.

On the regional calendar, the thirty-third Southeast Asian Games were held in Thailand from 9 to 20 December 2026. The Asian Games in Aichi-Nagoya, Japan, ran from 19 September to 4 October 2026 and had just closed when this file arrived. Ahead lie the thirty-fourth Southeast Asian Games in Malaysia, expected in 2027, the World Athletics Championships in Beijing, also in 2027, and the Los Angeles Olympic Games in 2028. That is a clear four-year spine, planned long in advance.

Yet when I tried to fill in the very first cell, comparing a performance against world, Olympic, continental and national records, I had nothing to enter. Not from laziness. The marks in my hands were missing three minimum facts required for any comparison: wind reading, venue altitude, and declared shoe model. Those are not minor details. They determine the value of every comparison.

Wind, altitude, and the price of a hundredth of a second

World athletics regulations have long set the assisted-wind limit at 2.0 metres per second for sprint events under 200 metres and for jumps. Above that threshold, a mark still counts as a competition result but cannot be considered a record.

That 2.0 figure sounds technical, but its consequences are large. A tailwind at exactly 2.0 metres per second can help a male sprinter run the 100 metres roughly one tenth of a second faster than in still air, and the benefit for women is often slightly larger. One tenth of a second in the 100 metres is the gap between a semi-final and a final at a continental championship.

What matters more is that many regional meets publish no wind reading at all. Without one, a 10.90 and an 11.05 become incomparable, even though they sit one hundredth of a second apart on the sheet. Fans read the results, see who was faster, and draw a conclusion. That conclusion can be entirely wrong.

Altitude works the same way. Thinner air in cities such as Mexico City or Bogota reduces drag, and the benefit is clearest in sprints, the 200 metres, and the long jump. Bob Beamon's 8.90 metres at the 2026 Mexico City Olympics stood for twenty-three years, and nobody disputes that the 2,240-metre altitude contributed a substantial share of that record.

For Southeast Asian athletics, this has an extra layer. Most stadiums in the region sit near sea level, so regional marks gain almost nothing from altitude. But many national teams, Vietnam included, regularly train in Kunming, a city at roughly 1,890 metres. Training high and competing at sea level is sound physiology. The problem lies elsewhere: when an athlete sets a personal best at an altitude meet, that mark needs a discount before it is compared with a sea-level performance. Almost nobody applies the discount.

I remember sitting with a Chinese coach in the stands at a Kunming training track. He pointed at the lane and said something I wrote in my notebook: "Here everybody runs better than themselves, but not because they are better." That is true of training marks. It is also true of competition marks recorded at altitude.

The equipment dividend and the problem of comparing across eras

In 2026, world athletics authorities introduced the first limits on sole thickness and plate structure, capping road shoes at 40 millimetres. The rule arrived after a surge of performances tied to carbon-plated shoes, and it conceded something the industry had been reluctant to say out loud: footwear is a measurable variable.

Independent studies in the years since have estimated the energy savings of the new generation of shoes at a few percent, varying by body type, distance and running mechanics. A few percent sounds small. Over 10,000 metres it equals twenty to thirty seconds. Over a marathon, it is the distance between a medal and tenth place.

Beyond shoes there is the track surface. Several tracks rebuilt over the past decade allow better elastic return, and major championships tend to be staged on such surfaces. The Tokyo Olympics in 2026 is the most cited example, producing a wave of season's bests and national records.

The analytical consequence is clear. Comparing a national record set in 2026 with one set in 2026 without a note on shoe generation and track surface is a faulty comparison. Not because the 2026 athlete was weaker, but because they ran in a different equipment context. When I write about athletics, I set one rule for myself: every cross-era comparison carries a line about equipment. Without that line, I do not write it.

There is another layer rarely discussed. This technology did not stay in professional hands. Thick-soled carbon-plated models reached amateur runners very early. That means age-group records, mass-participation records and community race results across Southeast Asia are affected too. A junior age-group record set in 2026 and one set in 2026 were almost certainly not measured with the same ruler.

Small samples and the habit of reading one race as one person

Based on my experience following athletics competitions over many years, I set a minimum threshold before calling any mark an athlete's true level: at least three marks within eighteen months, at least two under legal wind, and at least one set in a final or in a genuinely competitive meet.

That threshold is not perfectionism. It comes from a simple biological fact: sprinters typically vary between 1.5 and 3 percent from race to race within a season. Over 100 metres, 3 percent of 10.50 seconds is more than three tenths. In middle-distance events, the absolute time variation is smaller but is dominated by tactics: a slow heat and a fast final can differ by five to seven seconds over 1,500 metres even when the athlete is in identical condition.

My own profession makes this mistake constantly. An athlete sets a personal best, and the coverage calls it a "new level". Three weeks later the same athlete finishes fifth at a major meet, and the coverage calls it a "decline". The truth sits in between and is less attractive: the athlete was always at that level, and simply produced one good race and one ordinary one.

This is the point I want to make clearly, because it is the centre of this piece. Numbers tell only half the story; the other half lives in trembling calves on the infield grass. When we take one race and use it to define a person, we discard the hardest part of this sport: repetition. Repetition is what separates a talent from an athlete.

The economy of unratified marks

There is a category of data in wide circulation that no governing body recognises: training marks.

An athlete runs a certain distance in training, a coach records it, someone photographs the stopwatch, and within hours the information spreads through social media groups. Headlines like "athlete X breaks a world record in training" appear regularly.

Training marks cannot be ratified for a series of technical reasons that cannot be skipped: timing devices are not certified to competition standards; there is no photo-finish system; no reaction-time sensor; no wind gauge; no doping control; and the training track is not necessarily measured to standard. Any one of these is enough to void a mark.

But the damage is not technical. The damage is that expectation is built from something that was never a competition result. When that athlete enters a real meet and runs at their actual level, the audience feels cheated. The athlete is criticised for a number someone else created.

I have watched this happen many times, on both sides of the border. Working in Shanghai and following Chinese sports media groups shows me how fast this information travels and how hard it is to correct. A false claim spreads in two hours; a correct retraction needs two weeks to reach the same audience.

This is where I check myself. Had I never lived in Vietnam, would I see this problem as clearly? Probably not. I might have treated it as a natural feature of a young media market. Having lived in both places makes me recognise that this is not a problem of one country but of every sporting culture where data has not yet been institutionalised.

Meanwhile, agents and negotiators understand the value of unratified marks very well. Every transfer contract is a small migration, carrying expectation and homesickness. In athletics, the most common form of "transfer" is recruiting young athletes from the provinces, or naturalising athletes for regional events. In both cases the negotiation file always contains a section listing marks that cannot be verified. The buyer knows it. The seller knows it. And the third party, the audience, is the only one who does not.

Missing splits and the cost of reading only the finish

There is another kind of empty cell that bothers me more than any other: splits.

A split is a time recorded at an intermediate point. In the 400 metres, the most important one is the first 200. In the 800 metres, it is the first 400. Over longer distances, you need every kilometre and the final 400.

Without splits, an analyst can read only the final result, and the final result conceals the shape of the race. Take a concrete example. Two athletes both finish a 400 metres in 45.80. The first goes through 200 metres in 21.50 and dies at the end. The second goes through in 22.30 and finishes faster. On the results sheet they are identical. In reality they sit on entirely different trajectories: the first has room to run faster with better distribution, the second has most likely hit his ceiling.

At 800 metres the gap is wider. An athlete who splits 50 seconds and finishes in 1:46 has completely different potential from one who splits 52 and finishes in the same time. The first needs to learn distribution. The second needs to raise base speed. Two opposite training directions, decided by a set of numbers nobody published.

Based on my experience watching races and meets, most athletics analysis in Southeast Asia is written without splits. The result is that we routinely misjudge athletes simply because they lost a race they ran correctly.

I remember one such case. A young middle-distance runner finished fourth in a regional final. The results sheet said fourth. The coverage said "failure". Only later, when I obtained splits from an amateur cameraman in the stands, did I see that the athlete had run the final 400 faster than anyone in the race, and that with twenty more metres the placing would have changed. I rewrote my piece. But the first version had already been read by more people than the second.

Risk landscape: four empty cells you can fill, one you cannot

Working through the risk section of the template, I found the risks divide into five groups, and four of them can be handled with method.

The first is interpretive risk: wind, altitude, equipment and small samples. All four are measurable and adjustable. They become risks only when ignored.

The second is data risk: missing data, late data, and inconsistent data across meets. This can be handled with a shared publication standard.

The third is regulatory risk, including anti-doping violations that can retroactively change historical results tables. This risk is routinely underestimated in regional analysis, even though it can erase a medal years later. Any historical ranking should be read with a note about the possibility of revision.

The fourth is market and career risk: an athlete mispriced in the sponsorship market because a mark did not represent their true level, or, conversely, a good athlete priced low because no data was ever published.

The fifth is systemic risk, and it is the only one method cannot fix. It belongs to the governance structure of the sport.

The contrarian view: stop sanitising competition conditions

Here I want to offer a view I know will irritate more than a few colleagues in data analysis.

Our instinct when we see a mark is to strip it of conditions: subtract the wind, adjust for altitude, discount the equipment dividend, discard outlier races. We want to find an athlete's "true level", a pure number independent of circumstance.

I think that goal is an ontological mistake.

A race is not a measurement taken under ideal conditions. A race is an event that happens under specific conditions, and those conditions are part of the event. The wind on the home straight, the track heating up after three hours of sun, the roar of the crowd at the final bend: these are not impurities to be filtered out. They are the content.

Tactics never die; they are only misunderstood until someone is brave enough to try them again. The correct way to run into a headwind is a tactical decision, and it differs from the correct way to run in still air. An athlete who sets a personal best into a headwind has done something harder than one who matched it with a tailwind. If we discount every condition to reach a neutral figure, we erase the very thing that gives the performance meaning.

The right approach, I think, is the reverse: instead of sanitising conditions, publish them as part of the result. A mark should read as a full sentence with a subject, a verb and an adverbial. For three decades we have been cutting off the adverbial and keeping the number, then wondering why readers no longer trust the number.

There is a second contrarian angle, about the politics of data. Empty cells do not exist purely because of technical incapacity. In many places, data is a bargaining asset. A federation that publishes full detailed results hands rivals and media the ability to evaluate it. A timing company holds raw data because that is a contract advantage. An agent holds training marks because an unverified mark can negotiate better than a verified one.

In other words, transparency in regional athletics is not a technical problem. It is a negotiation problem. And until someone has enough interest to break that equilibrium, the empty cells will stay.

The regional medal as a media product

There is a structural paradox rarely discussed in Southeast Asian athletics.

The Southeast Asian Games is the biggest investment target for most countries in the region, Vietnam included. A medal there carries enormous media and social value. But the path to world championships runs through a different system: direct entry standards set by the world governing body, or a points ranking accumulated across international meets throughout the cycle.

The consequence is a gap between two arenas. In some events, the mark that wins a regional gold still sits below the entry standard for the world championships. That makes peaking for the regional Games a sensible communications choice but a limited ranking choice.

I do not think that choice is wrong. For an athlete from a small province, a regional gold can change her entire life, and that is entirely deserved. But analysts should not confuse the two rulers. A regional medal measures the ability to win in one arena. The world ranking measures level against the globe. Both are real, and neither replaces the other.

An example I often use with younger colleagues: a Vietnamese athlete such as Nguyen Thi Oanh has repeatedly won Southeast Asian Games gold in the 1,500 metres, the 3,000 metres steeplechase and the 5,000 metres. The value of those medals to Vietnamese sport is undeniable. But anyone using regional medal counts to infer prospects at a world championship is using the wrong ruler. Two different questions require two different data sets.

What will fill the empty cells

Over the past eighteen months I have joined three regional athletics data projects, and all three failed at the same point: nobody would release raw data.

The first ended with a spreadsheet seventy percent empty. The second ended with a report drawing only on data already published by the continental federation. The third ended as an article with no tables at all.

What I learned is this: if you want athletics data in this region, someone has to create it, and it has to be done at the scale of one meet, not one system. Thirty athletes, one data recorder in the stands, and a results file published within twenty-four hours.

The minimum data set for each sprint performance includes: reaction time, splits, wind measured at the track, venue altitude, temperature and humidity, declared shoe model, and the ratification status of the result. For middle and long distance, add lap times and the final 400. That is about twelve fields. One person can record them in an evening.

I am not calling for a technology revolution. I am saying the improvement threshold here is very low, and the gap between having data and having none is not about money. It is about the will of whoever organises a meet.

I once built a spreadsheet for the World Cup before I learned that the stands never appear in the formula. I still believe in spreadsheets. But I believe more in recording the full conditions of a race, because that lets me respect both halves of the story: the part that can be measured and the part that cannot.

What is worth keeping for next season

Looking down the four-year spine toward Los Angeles 2028, I think what Southeast Asian athletics lacks most is not performances, not talent, and not money. What it lacks most is a record that can be argued with.

A medal cannot be argued with. A mark set under unclear conditions cannot be argued with either, in both directions: nobody can prove it right, and nobody can prove it wrong. A sport where every performance sits in that suspended state cannot accumulate knowledge between generations. Each generation of athletes has to start from zero.

At forty-two, I have learned that every athlete is a poet who never gets published. They write with their bodies, they write fast, they write in the wind, and most of what they write is never recorded. An evaluation system for athletes is no better than the data system that produces it. If we measure five data points to the hundredth and discard twenty data points about conditions, we are using a microscope on a mural.

When a young athlete from this region steps onto the track at the next Asian Games, data about her will be collected by the host's systems, stored on an international federation's servers, and resold to media companies. She will own none of those rows. Fans at home will receive four columns: name, country, mark, place.

What I most want from the next cycle is not a new national record. What I want is for one national federation in this region to publish splits within twenty-four hours of a domestic meet, in full, without selection, including the last finisher's race.

When that happens, I will believe that Southeast Asian athletics has begun writing its own book instead of letting others write it.

Until then, if you see an analytical sheet filled with N/A, do not assume the writer failed. Consider that the writer may have been the most honest person in the newsroom that day.