Wind, Altitude and Shoes: How to Read an Athletics Result Correctly

Three numbers sat side by side on the results sheet I was holding that day: +1.9 metres per second, 10.35 seconds, and 21 years old. The stadium announcer read exactly one of them. The crowd nodded, a few people raised phones to photograph the board and posted a short caption. Nobody mentioned +1.9. Nobody mentioned the age. I was standing in the technical area, thirty metres past the finish, next to the wind gauge, watching the needle jump.

That needle moves throughout the race. The rules only take a short-window average along the final straight, but wind does not blow evenly. In some races the first half is a crosswind and the second half a tailwind. In some races the athlete in the inside lane meets different air than the one in the outside lane, even though both run the same straight. The final number is therefore an average of many things that were never alike.

By the rulebook, +1.9 m/s is legal. It sits right against the +2.0 ceiling. The mark is ratified, filed, counted in every national and regional ranking. By the data, it is a number that received almost the maximum assistance the rules allow, and anyone comparing it with a windless race has to do the conversion by hand. The scoreboard will not do it for you.

When the numbers speak, I simply listen.

The gap between a 100-metre race with a +1.9 tailwind and the same athlete running into a -1.2 headwind can reach two to three tenths of a second. Over 100 metres, two tenths is the entire distance between a place in the final and a stop in the semi-final. Same legs, same training load, same potential medal — only the wind direction differs, and the fate of an entire season flips.

None of this means the mark was fake. It means that comparing two results requires bringing them into the same frame of reference. That is the first principle, and the one most often skipped in online arguments after every athletics meet.

The four adjustment layers the scoreboard hides

The first layer is wind. In sprints and long jump, a tailwind is a free subsidy. That subsidy is not perfectly linear, but as a rough guide each metre per second of tailwind is worth roughly five to ten hundredths of a second over 100 metres, depending on the athlete and the acceleration phase. An athlete who runs 10.35 with +1.9 wind may run 10.60 in still air. Both numbers are true. Only one of them can be used for comparison.

The second layer is altitude. Above a thousand metres, the air thins, drag falls, and sprints, long jump and triple jump gain a visible dividend. Vietnam has a particularity worth remembering here: almost all of our tracks sit at sea level. National Stadium in Hanoi, Thong Nhat in Ho Chi Minh City, the tracks in Da Nang, Nha Trang, Can Tho — none of them pays an altitude dividend. That makes domestic marks hard to inflate on this count, but it also means that every trip to a high-altitude international meet requires a change of reading method.

The third layer is the track surface. Synthetic tracks are not a uniform material. A newly laid track, up to standard, with good rebound, can be several percent faster than one that has served ten years. Over 400 metres, several percent is more than a second — the difference between the podium and outside it. Domestic results carry the same meet names year after year, but the surfaces do not.

The fourth layer is the shoe. In recent years the technology race has moved from recreational running onto the competition track. Thick stacks, carbon plates, ultra-light responsive foams all create a percentage advantage. International federations have had to set stack-height limits, and for competition track shoes those limits were tightened well below road-shoe allowances. Domestically, the gap between the athlete with the best shoes and the one without remains a real variable. It does not appear on the scoreboard.

The performance curve: the hardest thing to fake

If I could keep only one class of data on an athlete, I would keep the year-by-year personal best curve. Not the personal best. Not the season's best. The whole series, placed on a timeline, one year at a time.

A normally developing athlete improves gradually, a few percent a year, with plateaus and occasional leaps. When one leap runs far beyond that athlete's own historical annual gain, it is a signal that needs checking. Checking, not convicting. The difference between those two verbs is the entire professional ethics of a data analyst. A leap can come from a new training plan, from moving to full-time training, from finally healing a nagging injury, from a coaching change, or from a source that is not permitted. My pencil does not know the answer. It only knows how to ask.

Distance never lies; we simply have not been patient enough to listen.

Then there is the age axis. Every event group has a peak window, and that window shifts with gender, physique and injury history. A 21-year-old sprinter running 10.35 sits at the start of that window. A 31-year-old running 10.35 sits at the end. The same number tells two entirely different stories: one of prospect, one of completion. Reading a result without reading the age is reading half a sentence.

I do not believe in luck; I believe in what has been repeated enough times.

And here is a rule I set for myself long ago: never diagnose an athlete's form after a single race. Two races minimum, three preferably, in comparable conditions. One poor race can be a mild illness, bad sleep, poor rhythm, wind, a track still damp after rain. One excellent race can be wind, a rival setting the pace, or plain adrenaline. Both are noise until they repeat.

Getting into the meet: two doors and one ceiling

The international system offers athletes two channels: hitting a qualifying standard inside a fixed window, or accumulating enough world-ranking points. Two roads, two strategies, two completely different ways of spending physical reserves. The first demands one perfectly timed race and turns every meet in the window into a gamble on when to unleash. The second demands steady presence, many meets, points in every round, and acceptance that each outing wears you down.

Then comes the ceiling. Each country may enter only a limited number of athletes per event. The consequence, in countries with real depth, is that a domestic fourth-place finisher can be faster than the champion of many other nations and still stay home. That is a structural risk invisible to individual data, and it only appears when you place a national ranking list beside an entry list.

For Vietnamese athletics, that ceiling is rarely reached in the flagship events, but it is a genuine variable wherever two or three athletes are level. A regional championship slot is sometimes decided by one race on one day, and the loser has not actually been weaker all season.

The landscape map: look at structure, not medals

World athletics has a fairly legible structure. Sprints cluster in a handful of countries with school-to-elite pipelines. Distance events cluster on the East African highlands. Jumps and throws belong to North America and Europe. Race walking and several women's throwing events carry a strong Asian presence.

In Southeast Asia the structure is similarly legible: sprints and relays belong to one group of nations, endurance and race walking to another, pole vault and hurdles to individual names. Vietnamese athletics, judged on multi-season data, is strong in middle-distance endurance and in events that reward a long-built physical base. Nguyen Thi Oanh is the clearest example of a multi-event athlete winning medals across several distances at the same Games. Nguyen Van Lai once held high positions in the men's 5,000 and 10,000 metres. Bui Thi Thu Thao left her mark in the women's long jump.

But the structure that matters most is depth. A country with three elite athletes and a large gap behind them is placing a bet every championship cycle on whether those three hold up. The cheapest test is the age structure of the leading group: if the five best athletes sit inside a narrow age band, a generational handover is approaching, and it will arrive all at once.

Rules and anti-doping: an empty dataset is not a clean dataset

This is the most misread section, so I will say it plainly: the absence of information is not evidence of innocence. When I open a dataset and find no row related to testing, that is an empty cell. An empty cell means unassessed, not confirmed safe.

Modern anti-doping runs on several pillars. Athlete biological passports are monitored over years, and unusual fluctuations in blood markers can open an investigation even with no positive sample. Whereabouts reporting and out-of-competition testing create a permanent net. Samples are stored for years and old samples can be re-analysed with new technology, leading to medal reallocation at championships that ended long ago.

For the reader of results, the practical consequence is that a medal table can change years later, and this is not rare. A medal awarded this season may be reconfirmed on a very distant date. That knowledge should not turn every result into suspicion, but it should make us cautious about permanent claims.

Alongside sits the technical rulebook, which quietly decides more outcomes than people assume. A false start means immediate disqualification with no second chance. Leaving your lane means disqualification. A relay exchange outside the zone means disqualification. In jumps and throws, three consecutive fouls mean no mark at all. An athlete can arrive with the best base in the field and go home empty-handed because of one technical error at the decisive moment.

Every number is a confession the competition cannot deny.

The training system behind every curve

No performance curve generates itself. Behind it there is always a structure: a coach, a training group, facilities, nutrition, recovery, sports medicine.

Training models differ fundamentally around the world. Some athletes grow up inside a school pipeline, compete year-round, and turn professional as a matter of course. Some train in national centres from a very young age. Some come from an entire highland belt that functions as a natural academy, where children run to school every day and the fastest of them become internationals.

Vietnam follows the centralised model, with sports talent schools at the base and national training centres at the top. The model concentrates resources and delivers results in priority events. Its weakness is dependence on a few key individuals: one good coach, one good doctor, one good analyst. When one of them leaves, a whole cohort of athletes can decline within one or two seasons.

Testing this risk needs no complicated data. Just place the performance history of a group beside the staffing timeline of the centre. If the two curves change direction at the same moment, you have a hypothesis to test.

The risk called a training mark

There is a class of number that appears regularly in the press and almost never carries comparative value: a mark measured in training, or hand-timed, or recorded at an internal test with no officiating.

Three technical problems. First, hand timing always carries error, and the error usually leans in favour of the person holding the watch. Second, training conditions are rarely recorded: wind, temperature, surface, time of day, prior training load. Third, with no finish-line image, there is no way to verify.

That does not make all training marks meaningless. For coaches they are valuable data for adjusting plans. For the media they are unverified data, and the only honest way to use them is to label them clearly.

The contrarian angle: correlation is not causation

This is where I regularly have to tell myself to slow down. Sports data is full of variable pairs that move together, and the human brain is built to wire them into cause and effect immediately.

An athlete switches shoes and breaks through. The story told is about the shoes. But over the same period, that athlete's fast-running volume may have risen, a heel injury may have healed, or the athlete may simply have entered the hinge year of the age curve. The shoes are just a variable sitting next to another variable on the timeline.

A team changes coach and results improve. The story told is about the new teacher. But the team may also have changed its competition schedule, cut the number of meets, or moved to a training camp with a more suitable climate.

At minute 70, the crowd sees collapse; I see a structure being rebuilt.

And the reverse holds just as firmly. One good race is not a new performance tier. One bad race is not the end. Both need repetition before they become conclusions. My job, in the end, is to wait patiently for enough repetitions before saying anything at all.

People ask why I stay silent; I am reading the lines the track writes.

Looking to the next cycle: five signals to watch

The coming cycle will answer questions the current data cannot.

The first signal is wind speed recorded and published in full at domestic meets. Once every sprint and long jump carries a wind reading, comparisons become automatically fairer and pointless online disputes shrink considerably.

The second signal is split-time data. A 400-metre race can be run many ways, and pacing distribution says more about fitness, tactics and maturity than the final time. When domestic meets publish splits, readers gain an entirely new layer of understanding.

The third signal is the age structure of the leading group in each event. If the top three in an event are all over thirty, that is information about that event's near future, and it deserves to be written a cycle earlier.

The fourth signal is the appearance of overseas competitions in the calendars of leading athletes. The number of international meets per year is an index of competitive environment quality, and it usually predicts regional results better than any fitness test.

The fifth signal is whether training marks are labelled. One small line stating the measurement conditions would lift the information quality of the whole sport by a level, at almost no cost.

At the end of every meet, when the scoreboard lights are off and the stands have emptied, I walk a lap of the track. Once a young athlete sat at the edge of the field, shoes still laced, and asked a question I still remember: "Brother, was my result today real?" I looked back at the results sheet, found the wind row, and told him it was real, it just needed to be read alongside the other numbers. He nodded and bent down to untie his shoes. I think that is the whole of my work: keeping every number read alongside the others, so that nobody has to ask whether their own result was real.