Nine Dimensions and One Empty Cell: A Protocol for Reading Athletics Before Concluding

Three in the morning in Osaka. My junior colleague slides a nine-page report across the desk. He followed the procedure correctly: he isolated the event and classified the mark; he reconstructed the personal-best curve; he checked the qualifying standards and the world ranking; he drew the competitive map of the discipline; he reviewed the rules framework and the anti-doping layer; he examined the training system; he built the six-category risk matrix; he analysed the media narrative and the expectation cycle; and finally he diagrammed the transmission into the market. Nine dimensions. Nine pages. Every page with tables, boxes, arrows and source notes. In the conclusion cell of all nine pages, the same line: Insufficient information – cannot be assessed. He asked me: So what is the conclusion? It took me forty minutes and two rounds of coffee to answer a question that is four words long in Vietnamese: cannot conclude yet. That is the correct answer. And in this trade, it is the most expensive answer, because it cannot be sold to anyone. Athletics is the strangest sport in the competitive system. Every claim in it collapses into three units nobody can argue with: milliseconds, centimetres and heartbeats. There is no contested sideline, no three-week refereeing dispute, no twelve-metre shot or 0.04 xG to fight over. One athlete runs 9.79 and the other runs 9.80. There is no interpretation. That cleanliness creates the illusion that a number is a conclusion. That illusion is the root of nearly every mistake I have made in this profession. Athletics data looks like an audited spreadsheet, but it is really a stack of conditions: wind, altitude, surface, shoe model, split distribution, field quality, round, and position in the season. Strip those away and what remains is a character, not a number. The nine-dimension protocol my colleague used was not born in a research institute. It was born on a betting desk in Osaka, where being wrong costs real money, and where you are forced to write down a rule that sounds trivial: if there is no data, write that there is no data. That rule is called null handling. The sports analytics industry hates it. A report with nine empty cells cannot be presented to a client. A report with nine numbers can be presented even when all nine are guesses dressed in formatting. Data never lies; the liar is the person who chooses how to read it. But that sentence only covers half the problem. There is a third and more dangerous kind of liar: the one who reads from a blank page and pretends the page is full. What makes an analytical file go empty? Not laziness. Usually it is a silent failure upstream – a document that failed to load, an extraction stage that broke, data truncated before it reached the analyst. Silent failure is the most dangerous failure class in data work, because it raises no alarm. It returns a tidy, correctly formatted, entirely meaningless result. So when I looked at that nine-page report, my first question was not about content. My first question was: does the source document exist, and does it contain words. Based on my experience of tracking athletics events across Japan and Southeast Asia for more than two decades, I have seen the same scenario many times. An expert receives a table of numbers, builds a beautiful analytical structure, and publishes a conclusion without first checking whether the table contains anything. A beautiful structure always beats an empty truth in the reader's eyes. The nine dimensions below are the protocol I use before allowing myself to conclude anything about a track. The order matters less than the discipline: each dimension must either change a decision or change a perception. If it changes neither, it is decoration. Dimension one: the stamp on the track Before comparing any mark to a record, classify it. Athletics contains at least five categories of mark that differ in legal and predictive value: officially ratified, wind-assisted, altitude-assisted, indoor, and unratified training marks. The wind threshold is one of the most misread numbers in the sport. In sprint and jump events, a mark counts for record purposes only when the tailwind does not exceed 2.0 metres per second. Above that threshold, the mark still signals potential, but it does not signal class. A 10.05 run into a 2.3 metre-per-second tailwind in April is not the same asset as a 10.05 run into a 0.4 metre-per-second headwind in August. Two lines of data, identical digits, different everything else. Altitude is the second variable. At stadiums roughly a thousand metres above sea level, thinner air helps sprints and jumps while hurting long-distance events. This is why certain training hubs in East and Southern Africa became year-round gathering points, and why marks recorded there must be read with a different ruler. Shoes are the third variable, and the most underrated. Since 2026 the World Athletics rules have set a baseline sole-thickness limit: a maximum of 40 millimetres for road shoes and 20 millimetres for track spikes, alongside limits on carbon-fibre plate construction. Those rules arrived after a period in which world marks across several distances climbed fast enough that the governing body itself had to concede that midsole technology had become a variable inseparable from the result. Finally there is split data. A sprint result without splits is a mute mark. Two athletes can both run 9.95, one by hitting maximum velocity in the first thirty metres and holding, the other by distributing evenly and unleashing over the last twenty. Those two patterns forecast two different futures once a three-round championship is in play. Dimension two: the curve of a body A personal best is a point, not a line. The analyst's job is to reconstruct the line. An athlete whose best improves steadily by a few percent a year is a different case in kind from one who treads water for four seasons and then jumps a whole level at once. The age curve depends on the event, and this is where outsiders err most. In sprint events the peak usually falls between twenty-four and twenty-nine. In the marathon the peak can extend past thirty-five, sometimes well past. The same date of birth cannot support the same conclusion for two athletes in different events. There is one quantitative tripwire I use as a relay: if a single season's improvement exceeds roughly three times that athlete's own historical annual gain, the file must be pushed into cross-validation, together with competition history, testing frequency and biological data where available. That threshold says nothing about ethics. It says the probability of an unmodelled cause has risen. Recovery is never a miracle; it is only something you already saw in the numbers three months earlier. An athlete returning from injury and running faster than the old baseline has usually already shown the signal in internal testing, in training volume, in range of motion, long before the public saw the result. Another common error is mistaking the best race of a life for a new level. A sample of one race is not enough to upgrade a profile. Upgrading requires at least three repetitions under comparable conditions. Dimension three: the narrowest door Qualification is where data becomes entitlement. The current World Athletics system for the Olympic Games runs on two parallel paths: hitting the entry standard, or accumulating enough world ranking points. Each event has a total quota and a maximum of three entries per country. Take the Paris 2026 cycle as a reference point: the standard in the men's 100 metres was 10.00 seconds, in the women's 100 metres 11.07, in the men's marathon 2 hours 08 minutes 10 seconds, and in the women's marathon 2 hours 26 minutes 50 seconds. These numbers are not absolute measures of class; they are measures of competitive density in a given event at a given moment. The three-per-country cap creates an effect I call internal involution. In a country with great depth, the fourth-ranked athlete domestically may have a better mark than another country's champion and still miss the Games. Without seeing this mechanism, an analyst reads a start list as a ranking of strength, when it is only the output of a quota algorithm. The time window is the final and most easily ignored variable. A mark run outside the valid window is worth zero for qualification purposes, even if it is the best of a career. An athlete who hits the standard in June and does not race again until the Games carries two layers of risk: injury risk and loss of competitive rhythm. Dimension four: the power map of an event Every athletics event has its own shape. Some events are dominated absolutely by one person for years. Some are two-horse races. Some are chaotic with six or seven evenly matched contenders. And some are in generational transition, where the outgoing generation has not fully left and the incoming one has already arrived. That shape governs how every number inside it should be read. A mark in a dominated event often reflects the tension of the race rather than a biological ceiling. In finals with a strong pacemaker, the whole field's average mark rises; in finals with nobody pushing, the average falls while the gaps compress. The same gold medal, two different meanings. When everyone is looking in one direction, I start examining the gap behind their backs. At national level, the athletics power map is not built by culture but by pipelines. Jamaica produces sprinters through a dense school-meet system. East African nations produce endurance through terrain, altitude and an economy in which running is a genuinely paid profession. The United States college system produces depth across nearly every discipline through scholarships and facilities. Japan, where I live, is a case worth studying on its own. The ekiden system, with the Hakone Ekiden covering roughly 217.1 kilometres across ten stages in early January each year, produces an endurance pipeline that is extraordinarily efficient in volume: Japan's count of marathoners under 2 hours 10 minutes is consistently among the world's highest. But that depth does not automatically convert into global medals, because the domestic competitive structure rewards year-round durability over a single-day peak. It is the cleanest example that depth and peak are two different indices measured with two different rulers. Dimension five: the grey zone of the rules The rules and anti-doping layer is where analysts most easily fool themselves, because it offers the feeling of ethics in place of the feeling of data. Four rule layers must be separated: technical competition rules, the anti-doping framework, eligibility regulations, and equipment regulations. These operate on four different timelines, and a file can be clean in one layer while snagged in another. The Athlete Biological Passport, in use since 2026, monitors biological markers longitudinally. Its value lies in detecting anomalies a single test would miss. On the whereabouts side, three missed tests or three filing failures within twelve months constitute a violation, even with no adverse analytical finding at all. In eligibility, the regulations on female athletes with differences of sex development, issued in 2026, set a testosterone threshold of 5 nanomoles per litre for events from 400 metres to one mile. Those regulations have passed through multiple levels of adjudication, including a ruling by the Court of Arbitration for Sport, and remain one of the most legally contested zones in the modern sport. In equipment, the sole-thickness limits described in dimension one are not merely technical. They are a continuous negotiation between manufacturers, federations and athletes, and each adjustment changes the relative value of every mark recorded before it. One principle I apply absolutely: the absence of a doping signal in a source is not evidence of a clean profile. It is evidence of an absent source. Those are two different statements, and blending them is the fastest way for an analysis to become an advocacy document. Dimension six: the system behind the medal No athlete runs alone. Behind every mark sits a coach, a training group, a training base, a science team and a power structure. Training bases are a quantifiable variable. Altitude camps in Iten or Font Romeu are not cultural symbols; they are tools that change red blood cell volume, and their effects are measurable through physiological markers within weeks. A file built on altitude training during the preparation phase reads differently from one built entirely at sea level. In one month of 2026, while working as a data commentator for a Japanese sports broadcaster during a World Cup group-stage match, I mispronounced a midfielder's name three times in the first half. Viewers remember that error. But what kept me awake was something else entirely: tracking data showed the team's shape had been stretched to an average of 42 metres, breaking the pressing structure. Mispronouncing a name is not the error; the omission is failing to see the outline of a system. That lesson transfers directly to athletics. When an athlete declines, the right question is not whether they lost form. The right question is whether the training group changed coaches, whether the training base moved, whether the medical team lost a key person. Those changes usually appear months before results do, and they appear in small news items, never in the results table. Dimension seven: the empty cell as a risk The risk matrix has six categories: competitive, anti-doping, financial and career, rules and eligibility, public opinion and brand, and systemic. Each carries a probability, an impact and a mitigation. One thing must be stated plainly: when the entire input is empty, the largest risk does not sit in any of those six categories. It sits in the decision taken on that input. A conclusion built on a zero-evidence base is a conclusion that cannot be reproduced and cannot be audited. In my trade, that is the definition of a disaster. There is a reverse trap alongside it. Because the source contains no adverse signal, a hasty reader may treat the subject as risk-free. The correct reading is the opposite: an empty source does not confirm risk, but it does not remove it either. No dimension has been cleared; all of them have simply gone unexamined. Here I apply Occam's razor with discipline. If a surface explanation is sufficient to explain the phenomenon, I stop there. My profession carries an occupational temptation: always hunting for a deeper order behind the data. Most of the time that deeper order exists. But there are moments when a 10.05 is simply a 10.05, and inventing a system behind it is a data error, not a profundity. Dimension eight: the story that sells tickets Sports media runs on a heat cycle: germination, acceleration, climax, backlash. Each phase has its own narrative structure. There is the record-hunt story. The prodigy story. The national glory story. The comeback story. The farewell story. And the doping story. The analyst's job in this dimension is not to join the story but to measure the gap between market expectation and objective assessment. That gap is the most valuable variable of all, because it exists before the result does. The prodigy filter is mandatory. A seventeen-year-old running a very fast mark sits in a small-sample zone. The probability that a peak mark at seventeen repeats at twenty-two is not high, because the body is unfinished and training volume is still low. The crowd reads such a mark as a starting point; the data reads it as an outlier awaiting more samples. Another variable that media almost never prices into expectations is the medal reallocation queue. When an athlete is disqualified or sanctioned after an event, medals are reassigned in order, and the process can take years. At the London 2026 Olympic Games, the women's 400 metres hurdles was decided in one order and revised years later after a sanction, with the athlete promoted to the top step receiving her medal at a ceremony held nearly a decade after the race. For some nations, the medal table of a given Games can still move after those Games are over. The medal table is not a closed document. Dimension nine: the current leaving the stadium Athletics does not end at the finish line. It flows into technology, commerce, media and derivative markets. On the technology branch, the midsole race is governed by rules the federation sets and revises continuously. Each time a thickness limit changes, the relative value of every historically recorded mark changes with it. This is why cross-era comparison in athletics must always carry a technical footnote. On the commercial branch, the sport's income structure splits into two worlds. Official prize money at major international meets is far lower than the public imagines. At the Paris 2026 Olympic Games, World Athletics paid prize money for individual gold medals for the first time, at 50,000 US dollars per athlete, with relay teams splitting the equivalent of one share. Meanwhile, most of the real income of top athletes sits in appearance fees at major marathons, where a single contract can exceed an entire season of official prize money. Every movement in the odds is a heartbeat; I can only hear it with my ear pressed to the ground of data. But on this branch I keep a hard boundary: analysing market structure is one thing, issuing betting recommendations is another. This analysis does not cross that line. Systemic risk sits in the final layer: the youth pipeline, national funding, and the health of the competition system itself. A country that cuts its athletics budget will not lose medals that season. It loses medals roughly eight to twelve years later, when the cohort that should have been developed during the cut no longer exists. It is the slowest-moving and least reversible variable in the sport. The counter-intuitive angle For years I believed that the more complete the analytical tooling, the more trustworthy the conclusion. I was wrong on one specific point: a beautiful structure can substitute for content. A nine-dimension table with full headers, columns, arrows and footnotes looks far more like professional analysis than a short paragraph saying there is not enough data. Readers cannot inspect the depth of a table. They can inspect its form. So in data work, the most serious error is not a wrong number. The most serious error is a report that passes the format check while carrying zero information. That is why I make every colleague read the conclusion line aloud before presenting the rest. If the conclusion cell is empty, everything else has value only as a process description. There is a fair counter-argument to my position: null discipline can become avoidance. An analyst can hide behind the phrase insufficient information to escape responsibility for any judgement at all. I accept that criticism, and it forces me to set a threshold: once the data is sufficient to change a decision, silence is a mistake, not a virtue. One more point I want to state directly, because I have made this mistake myself. Crowd emotion is not data garbage. It is raw data. It can be defined, measured and compared: article volume, ticket demand, discussion level, propagation speed. The way to handle it is neither contempt nor obedience. The way is to place it in a separate column with its own scale, and always compare it against the performance-fundamentals column. The ratio between those two columns is what is worth reading. Finally, something I must always remind myself of. An analyst has no right to judge someone else's reading without first publishing their own. I always place the opposing reading on the table before rejecting it, because otherwise I am not analysing data; I am presenting a bias in table format. What to watch in the next round Five signals will stay on my desk in the coming months. First, the null rate in my own reports. If that figure rises, the problem is not market data; it is the collection pipeline. A silently broken pipeline generates a stream of correctly formatted, semantically empty reports, and that is the most expensive class of error there is. Second, the condition labels attached to every mark: wind, altitude, shoe model, round. When a mark is published without those labels, its predictive value drops close to zero, however impressive the number looks. Third, the medal reallocation queue. It is slow-moving data, but it touches medal tables, sponsorship and history. Fourth, amendments to equipment regulations, because each amendment forces a re-reading of every historical comparison point. Fifth, the quality of the source at the input stage. An era does not begin with technology; it begins with a question sharp enough to cut through the rut. My question at this moment is very simple: does the source document contain words, and have I actually seen it. Every conclusion in this sport can wait. What cannot wait is a conclusion issued before the data arrives.