Vietnamese Athletics and the Unclosed Data Equation: From the SEA Games to the Olympic Start Line

VIETNAMESE ATHLETICS AND THE UNCLOSED DATA EQUATION: FROM THE SEA GAMES TO THE OLYMPIC START LINE

In mid-May 2026, on the My Dinh track, a female athlete crossed the finish line of the 1500m with a comfortable margin over the chasing pack. The stands erupted, red flags with yellow stars covering the grandstand. But what I carried back to Osaka was not that moment. What I carried back was an unanswered question: how many seconds faster was her final lap than the average of her first three, and what does that number say about her ability to compete on a different stage?

I sat in those stands with a notebook divided into four columns. Column one recorded cumulative time per lap. Column two recorded the relative gap between the leader and the second group. Column three recorded the athlete's physical state over the final 200 metres. Column four — the emptiest column — recorded what I could not measure from the stands. After four days of competition, column four was fuller than the other three combined. That is why this article exists.

Athletics is the sport where Vietnam's public data sits at the earliest stage of development compared with every team sport. We have football scores, we have basketball three-point metrics, we have volleyball blocking statistics. But we do not have a complete, public, standardised split-time system for sprint and middle-distance events. Without that data series, any analysis of Vietnamese athletics must begin by admitting this: we are judging athletes by the finish line, not by the process of reaching it.

CONTEXT: AN ATHLETICS NATION STANDING ACROSS TWO LAYERS OF COMPETITION

To discuss Vietnamese athletics, we must first separate two layers of competitive space, because they operate on entirely different logic.

The first layer is the Southeast Asian Games. At this layer, Vietnam is a genuine power. Across recent SEA Games editions, athletics has consistently been among the sports contributing the most gold medals to the Vietnamese delegation, regularly competing directly with Thailand and Indonesia for the top spot in the discipline. This is where names like Nguyen Thi Oanh, Nguyen Thi Huyen, Quach Thi Lan, Bui Thi Thu Thao, Le Tu Chinh and Nguyen Thi Thanh Phuc become national symbols.

The second layer is the continental and global stage. At this layer, the gap is no longer measured in medals but in seconds and centimetres. A SEA Games champion in the women's 400m hurdles may be three to five seconds behind an Olympic finalist in the same event. Five seconds over 400m hurdles corresponds to a physical distance the naked eye cannot read from the stands — but it corresponds to many years of systematic training.

The distance between these two layers is the real subject of this article. Not the story of medals, but the story of the data structure behind the medals.

I came to athletics by a roundabout route. In 2026, while still a school student, I recorded every match of the Japanese national team at the World Cup in Russia. On that Russian night in 2026, I watched data collapse before my eyes: Japan held 55 percent possession but touched the ball inside the opponent's penalty area only seven times, against twenty-one for Belgium. I wrote that pushing the defensive line high in the final minutes was a mistake. I was criticised. I held my position, because data does not lie — but I also learned that data only avoids lying when we measure the right thing.

In 2026, when the pandemic suspended competitions, I sat in Osaka and built a self-made dataset from old match footage, logging 1,240 pressing situations from Cerezo Osaka's 2026 season to calculate PPDA. When the league returned, I predicted Cerezo would decline because of the absence of home crowds. They finished fourth, lower than my second-place prediction. I admitted the error and added a "crowd influence" variable to the model. An empty stadium, yet the numbers were still full of noise.

Those two lessons followed me into athletics. First, measuring the right thing matters more than measuring a lot. Second, the variables left out are often more important than the variables included.

CORE: DISSECTING THE EVENTS WITH WHAT CAN BE MEASURED

1. The 400m hurdles: where rhythm data decides everything

The 400m hurdles is the strangest event in athletics. It is the only event where the result depends on a sequence of decisions repeated ten times, roughly thirty-five metres apart, each with an extremely narrow margin of error.

In this event, the most important variable is not top speed but the number of strides between hurdles. Elite female athletes typically take fifteen strides between hurdles for most of the race, then shift to sixteen or seventeen in the closing section as fatigue sets in. Every rhythm change costs speed and alters the take-off foot placement.

In the 400m hurdles, a three-centimetre error at hurdle take-off can multiply into a three-tenths-of-a-second error at the finish — and three tenths of a second is the distance between a medal and elimination.

Vietnam has Quach Thi Lan and Nguyen Thi Huyen, two athletes who have competed for years at the top of the Asian field, with medals at continental championships and Asian Games. What I want to emphasise is not their results, but the fact that we have almost no public stride-pattern data for them. We know the finish time. We do not know at which hurdle they shifted from fifteen to sixteen strides. We do not know how far their take-off foot placement deviated from the optimal point. We do not know their first three hundred metres rhythm relative to continental rivals.

That is the first and largest gap. A 400m hurdler can improve by half a second simply by optimising stride pattern, without increasing training volume. But to optimise stride pattern, the coach needs data. And that data must be collected in every session, at every hurdle, at high frequency.

In Japan, universities operate on-site measurement systems for each track athlete, including high-speed video analysis for every hurdle clearance in training. That is not expensive technology. It is process. A phone shooting at one hundred and twenty frames per second placed at a fixed angle, plus a person timing, is enough to generate valuable stride data. Our problem is not equipment. Our problem is establishing the habit of measurement.

1. The endurance cluster and the "one athlete, many medals" model

Nguyen Thi Oanh is the most interesting case in Vietnamese athletics over the past decade, and also the hardest case to analyse with purely numerical data.

She competes across overlapping events: 1500m, 3000m steeplechase, 5000m, sometimes 10000m. At SEA Games editions, she typically enters multiple events and wins gold in most of them, in some editions taking three to four individual golds. This is a rare achievement in any athletics nation, because endurance events demand recovery between competitions.

From a data perspective, three variables need measuring to understand this achievement.

The first variable is the actual recovery interval between finals. At the SEA Games, the schedule is often compressed to fit the duration of the Games. If an athlete runs a 3000m steeplechase final and a 1500m final less than forty-eight hours apart, winning gold in both reflects not only physical capacity but also the competitive depth of those events. In other words, gold medal count is an indicator contaminated by competition structure.

The second variable is speed distribution within each lap. An endurance athlete can win by running evenly, or by surging over the final four hundred metres. These two tactics demand entirely different training structures. The even-pace tactic requires a large aerobic base. The final-surge tactic requires high lactate tolerance. Without split data, we do not know what the athlete is winning with, and therefore do not know what needs improving.

The third variable is opponent quality. A SEA Games gold in the women's 5000m can be won at a time that would only qualify for the final at an Asian championship. When opponent quality changes, the same time carries an entirely different meaning — and failing to convert for opponent quality is the most common analytical error in Vietnamese sports media.

I want to be clear: I am not diminishing SEA Games achievements. I am saying that measuring achievement must come with measuring context. The gold medal is a fact. But "what the gold medal means" is a data question, not an emotional one.

1. Sprints: where reaction data decides

In sprint events, the first variable is not speed but reaction time at the start. In professional athletics, reaction-time measurement devices are fitted to the starting blocks and record the interval from the gun to the first force application. Under current World Athletics rules, an athlete is deemed to have false-started if reaction time falls below a certain threshold, and under the current format, any false start results in immediate disqualification.

This means that in the 100m, athletes must balance two opposing risks: starting too slowly loses the advantage, starting too quickly loses eligibility. In that brief window, the decision is made with no data feedback whatsoever.

In esports, human reaction is the limit of data. In sprinting, human reaction is the data — and it can be trained, but only to a certain biological threshold.

For Vietnamese athletics, Le Tu Chinh is the representative name in the sprint group. The analytical problem here is structural: improving 100m performance requires intervention in the acceleration phase, the maximum-velocity maintenance phase, and the deceleration phase. These three phases can be measured separately using camera systems along the track. Without measurement, a coach can only guess that the athlete needs to "run faster", an instruction with no technical value.

1. Jumps and throws: events lacking long-term data series

Bui Thi Thu Thao once won an Asian championship gold in the women's long jump, one of the highest achievements of Vietnamese athletics at continental level in the past decade.

Long jump is the event where data can intervene most deeply. Long jump performance depends on four components: approach run speed, accuracy of the final step, take-off force at the board, and in-flight efficiency. Each component is measurable.

The final step is the most sensitive component. If the final step is longer than the average step by a few centimetres, the athlete loses speed. If shorter, the athlete loses take-off force. In developed athletics nations, long jumpers have a fixed step count established for each runway, checked with markers in training. That is basic technique, not high technology.

In the throws group, data is even thinner. Vietnam has had athletes with good regional results, but a public long-term data series essentially does not exist.

1. Race walking: the event the media forgets

Nguyen Thi Thanh Phuc is the representative figure of Vietnamese race walking, having won SEA Games gold multiple times in long-distance walking events.

Race walking is a special event because the result depends not only on speed but on technical compliance. Judges have the power to warn and disqualify an athlete if the lead leg does not maintain contact with the ground, or if the support leg does not maintain the correct posture. This means a faster walker can be disqualified, while a slower walker with clean technique still wins a medal.

For data analysis, this is the event where the referee variable becomes the central variable. In Japan, a nation with a very strong race walking tradition, with names like Toshikazu Yamanishi and Koki Ikeda having won medals at world and Olympic level, technical training is conducted with video analysis from multiple angles, including rear-facing shots to check support-leg posture.

Vietnam has a foundation in this event at regional level. But to rise to continental level requires investment in detailed technical analysis, not only in training volume.

1. The data gap: three layers of deficit

Taken together, I see Vietnamese athletics lacking data at three layers.

The first layer is public competition data. Final results are published, but per-lap split times, start reaction times, and detailed competition conditions (wind speed, temperature, humidity) are often not fully published. This is the easiest layer to fix because it requires only recording and publishing processes.

The second layer is training data. This is the hardest layer because it touches athlete privacy and coach intellectual property. But even internally, many national teams still lack systematic recording habits.

The third layer is longitudinal data. An athlete must be tracked continuously across years to establish an individual development curve. This curve is the single most important tool for detecting anomalies, evaluating training effectiveness, and projecting career peaks.

Data does not create stories; it strips bare the stories of others. And in Vietnamese athletics, the story most stripped bare is the story of individual endurance substituting for a system.

CONTRARIAN: WHAT WOULD BE TRUE IF I AM WRONG

Before concluding, I want to argue the opposite case myself, because that is the rule I set after the 2026 season.

If I am wrong, the error lies in the assumption that Vietnamese athletics needs a detailed data system in order to progress. There is a strong counter-argument: in a phase of limited resources, concentrating on a few exceptional individuals with maximum support may be more effective than building data infrastructure for the entire system.

This argument is not without evidence. Many nations with strong athletics traditions passed through an "individual hero" phase before building a system. Japan had a period of reliance on exceptional marathon individuals before ekiden became a large-scale athlete production machine. Ethiopia and Kenya produced global stars before organised training centres existed.

If that argument holds, then Vietnam investing in a few priority athletes, with support teams including doctors, nutritionists and technical analysts, is the rational path. System-wide data can wait.

I think this argument is partly right, but misses one variable: transferability. When success depends on an individual, that success is not encoded into process. When that individual retires, the system returns to its starting point. Japan is not strong in marathon because of a few individuals; it is strong because it has a system producing hundreds of sub-2:10 runners every year. Ekiden, as a competitive institution, turned fast running into a cultural standard, and that standard is maintained by data: per-leg times, per-school rankings, head-to-head histories between schools.

So the real question is not "data or no data" but "data for whom". If data serves only one athlete, it is a tool. If data is published and standardised, it becomes infrastructure. The two cost almost the same but differ enormously in long-term value.

Another counter-intuitive point deserves mention: Vietnam's dominance in certain SEA Games events may be masking risk. When an athlete wins many golds, the pressure to maintain results rises, and that pressure typically leads to increasing training volume rather than improving recovery quality. This is the point where sports medical data must be tracked alongside performance data.

I collect mistakes, classify them, and then I know where the team is headed. With athletics, I do the same with injuries: every injury is a data point about the limits of the current training model.

TRAINING SYSTEMS AND YOUTH DEVELOPMENT: A STRUCTURAL COMPARISON

One of the questions I receive most from Vietnamese readers is: why does Japan produce so many world-class track and field athletes?

The short answer is: because Japan makes athletics part of the education system, not a specialised sport detached from it.

Ekiden is the clearest example. This is a form of long-distance relay racing between teams, most popular at high school and university level. The Hakone Ekiden, held in early January every year, is broadcast live with viewership ratings among the highest of any annual event in Japan. The important thing is not the race itself but its consequence: thousands of high school students train every year with the goal of making their school team, and hundreds of universities maintain systematic running programmes.

The result is an athlete population pyramid with a very wide base and a very high peak. In the men's marathon, Japan regularly has dozens of athletes running under 2 hours 10 minutes each year. That is not the achievement of a few individuals; it is the achievement of a system.

In other events, Japan has also made notable strides. In the 3000m steeplechase, Ryuji Miura set a national record and entered the world's leading group, an achievement Japan had never previously reached in this event. In race walking, Toshikazu Yamanishi and Koki Ikeda won medals at world championships and the Olympics, making Japan one of the strongest nations in this event group.

What is common to these successes? Not innate talent. It is identifying a target event, building a long-term pathway, and measuring continuously.

In the 3000m steeplechase, Japan recognised that this was an event dominated by East African nations but with lower global competitive density than the 5000m and 10000m. That was a strategic decision based on data analysis: choose an event with a competitive gap.

In race walking, Japan recognised that this event demands high technique, and technique can be systematically trained. That was also a strategic decision.

What can Vietnam learn here? Not to copy ekiden, because ekiden is tied to Japanese school culture and cannot be transferred wholesale. What can be learned is method: choose target events based on competitive-gap analysis, build multi-year pathways, and measure continuously.

For Vietnamese athletics, several events with continental competitive gaps include race walking, the throws, and certain jumps. These are places where competitive density in Southeast Asia and Asia is lower than in the sprints and middle-distance events — where nations with large populations and strong school sports systems hold the advantage.

COMPETITION RULES AND ANTI-DOPING: THE SILENT VARIABLES

In athletics analysis, there is a group of variables often overlooked in media coverage but with large effects on results: competition rules and eligibility regulations.

In long jump and triple jump, athletes have a set number of attempts, and any attempt beyond the take-off board is a foul with no mark. This means competition tactics include risk management: the athlete must balance precise foot placement against maximal take-off force. An athlete can jump farther than rivals in training yet lose in competition through three fouls.

In relay events, the exchange zone has strict limits, and exchanging outside the zone results in disqualification. This is a variable data can directly intervene in: exchange timing, the receiving runner's speed at handover, and the optimal distance between the two can all be measured and optimised.

On anti-doping, this is an area where I want to state a principle clearly: the absence of doping information does not mean the absence of doping risk. In data analysis, an empty result from an empty input is a meaningless result, not a clean result.

In athletics, control mechanisms include the athlete biological passport, long-term sample storage for retesting, and whereabouts rules for athletes in testing pools. These mechanisms create a data layer parallel to performance data, and this layer must be considered when evaluating any performance leap.

Every probability conceals a shock — I just make sure it does not repeat. In athletics, the way to ensure that is to track the individual development curve across years, because an abnormal leap always leaves traces in the data before it appears on the results board.

RISK AND THE VARIABLES THAT CANNOT BE MEASURED

This section is the one I consider most important, because it is the one I most often get wrong.

The first risk is small-sample risk. A gold medal at a single Games does not prove a stable level. In statistics, a single data point has very high variance and is insufficient to infer a trend. When media report a performance, they often present it as an established peak, whereas it may be a random fluctuation within a series of fluctuations.

The second risk is competition-condition risk. In sprint and jump events, wind speed directly affects performance, and under competition rules, a mark is only ratified as a record if wind speed falls within the permitted threshold. In middle- and long-distance events, temperature and humidity affect performance non-linearly. A mark achieved in cool conditions cannot be directly compared with a mark achieved in hot, humid conditions.

The third risk is equipment risk. Carbon-plated shoes have significantly changed performance in long-distance events over the past decade. This creates a variable that must be isolated when comparing performances across eras. Without isolating it, we may confuse physical progress with technological progress.

The fourth risk is missing-data risk. In many events, we have no split times, no condition data, no injury data. When data is missing, analytical models become sensitive to assumptions, and conclusions become less reliable. The correct handling is not to ignore the gap, but to state it explicitly.

The fifth risk is cultural-difference risk in analysis. This is the point I want to spend the most time on.

HOW I GOT IT WRONG WHEN APPLYING JAPANESE STANDARDS TO ANOTHER CONTEXT

In 2026, I spent three weeks following the Euros and wrote a piece on set pieces, comparing the effectiveness of designed set-piece drills at a Bundesliga club with the absence of similar structure at a certain Asian national team. The piece received about fifteen thousand reads and was reposted on a small football site.

But looking back, I see I made a methodological error. I assumed that a method effective in one context would be equally effective in another, without examining the underlying conditions: weekly training hours, support staff numbers, pitch quality, and even players' cultural habits.

I carried that error into athletics. When comparing Vietnamese athletics with Japanese athletics, I must constantly remind myself that the two contexts differ at very deep layers.

First is domestic competitive density. Japan has thousands of athletes competing in each event, and that density creates continuous pressure to improve. Vietnam has fewer athletes per event, and in some cases one athlete can dominate an event for years without a sufficiently strong domestic rival. This directly affects training motivation.

Second is competition structure. Japan has school, university and professional competition systems running year-round. Vietnam has fewer competitions, which means each outing carries greater psychological weight.

Third is measurement habit. In Japan, logging training numbers is part of coaching culture. In Vietnam, this is forming but not yet widespread.

Fourth is climate. Vietnam has hot, humid conditions year-round in many regions, which affects the ability to train at high intensity and affects competition tactics. A training model built in a temperate zone cannot be applied wholesale in a tropical zone.

I am not saying we should not learn. I am saying learning must come with context dissection. Otherwise we import a conclusion without importing the conditions that produced it.

WHAT WOULD CHANGE THE PICTURE

If I had to choose a single intervention with the largest impact on Vietnamese athletics over the next five years, I would not choose an athlete, nor a coach. I would choose a process.

That process has three steps.

Step one: standardise split-time recording at every final, in every domestic competition, and publish it publicly in machine-readable format. This is the lowest-cost, highest-value intervention. It creates a long-term data series any analyst can use.

Step two: establish long-term profiles for the priority athlete group, including individual performance curves, injury histories and competition histories. These profiles serve two purposes: training optimisation and anomaly detection.

Step three: build internal analytical capability within the coaching system, not outsourced. A coach who can read their own athlete's data is worth more than an analytical report delivered quarterly.

These three steps do not require expensive technology. They require discipline.

I know this because I once did the opposite. In 2026, analysing data on more than two hundred players moving from the Japanese domestic league to Europe, I focused on a single variable and found a correlation with statistical significance at 0.67. I got so excited about that number that I almost forgot correlation is not causation. A variable correlated with success may simply be a proxy for another unmeasured variable.

A contract is only the ending; the beginning is in the spreadsheet. With athletics, a medal is only the ending; the beginning is in the laps nobody counts.

TAKEAWAY: SIGNALS FOR THE NEXT CYCLE

What I will be tracking next season is not the medal table.

First, I am tracking how many domestic athletics finals publish split times. If that number rises, it is a signal that the system is shifting from measuring outcomes to measuring process.

Second, I am tracking the time gap between the SEA Games champion and the fourth-placed athlete in the same event. If that gap narrows, it is a signal that internal competitive density is rising, and competitive density is a precondition for progress at higher levels.

Third, I am tracking the number of young athletes entered in international competition under the age of twenty. This is an indicator of long-term investment.

Fourth, I am tracking whether a deliberately selected target event emerges. That selection, if based on competitive-gap analysis rather than tradition, would signal a new way of thinking.

Athletics is the sport where truth cannot be hidden by interpretation. The stopwatch does not argue. The tape measure has no bias. Therefore, every advance in this sport must begin with accepting the number, including the numbers that are not beautiful.

The question I carried back to Osaka after that SEA Games was not how many gold medals Vietnamese athletes can win. The question is: when will we have enough data to answer that question without guessing?