An Empty Column Is Not a Clean Chit: Auditing Indian Athletics Data

Last week a report landed on my desk with nine analytical dimensions and not one number in it. Title: N/A. Source: N/A. Athlete: N/A. Mark: N/A. The framework was fully built — PB progression curve, wind correction, altitude correction, qualifying standard, anti-doping checklist — and every cell held the one word no auditor ever wants to type: insufficient information. My first instinct was to blame the tool. My second, which usually turns out right: an empty column is not a zero. Zero is a measured value. An empty column is a question that was never asked. Treating the two as the same is the most expensive error in sports data, because a zero closes the argument and an empty column opens one — if anyone has the nerve to look at it empty. There was no athlete's name in that file, no event, no venue. A piece of athletics reporting in which all four of name, event, mark and venue are missing is not reporting. It is a broken pipeline. Indian athletics numbers travel through a narrow, badly maintained channel. There is no national league printing fresh data every weekend. AFI result PDFs arrive months late, agency copy is pasted from agency copy, and the official doing manual timing at the track is covering three events in one afternoon. Above all, there is no dedicated athletics beat that runs all year. Once every four years, for four weeks of an Olympics, the byline count triples; for the other 200 weeks the story is whatever the wire sends. In 2026 I scraped five years of AFI results. Three timing entries in the men's 400m final at the Federation Cup were manual and sat 0.3 to 0.5 seconds off the photo finish. I wrote it up as 'The 0.3-Second Lie', PTI cited it, and every project since has opened with a data audit: what is the source, what is the timestamp, how was it verified. August 2026, the Paris javelin final, me on the data desk. The live feed showed Neeraj Chopra's third throw as 89.45m twice and the graphic broke. I cross-checked the Omega timing system and the AFI manual log, confirmed 89.45, had the broadcast corrected, and logged Arshad Nadeem's 92.97m as the Olympic record. After that night the three-source rule became standard: primary sensor, official log, manual backup. No stat goes on air without a timestamp and a source code. That culture is what catches this empty file. The thread started with one number nobody had given. Now the part without which this report is incomplete: what happens to every dimension when the evidence base is empty. With no mark at all, there is no comparison to a world, Olympic, championship or national record, and none to a qualifying standard. The effect is bigger than it sounds: 10.15 seconds in a men's 100m could mean 10.00 or 10.35 on the same day if the wind reading is missing. Records count only with a tailwind of +2.0 m/s or less, and a venue above roughly 1000m helps sprints and jumps while penalising distance events. Without wind, altitude, track and shoe model, a 'true value' is an estimate, not a calculation. The second casualty is the PB progression curve. I run a simple test on every athlete: if a single year's improvement exceeds three times that athlete's historical annual average, check the wind, the altitude, the track and the shoes first — not the athlete. This file contained no year-by-year series at all. The third is the age curve. Sprinters peak around 24 to 29, middle distance 26 to 31, throws 28 to 33, and the marathon stretches past 35. Without a date of birth there is no positioning, and without an age there is no way to say where in a career the athlete in question stands. The fourth is the dual qualification channel. Entry to an Olympics or World Championships comes either through the direct qualifying standard or through World Ranking points. Calling an athlete 'safe' or 'borderline' without both is politics, not analysis. The national selection architecture matters just as much — a US-style one-race-decides-everything trials system can eliminate a reigning world record holder. The fifth is rules and anti-doping, where the most important sentence is this: the absence of red flags in this report is data absence, not a clean record. ABP status, whereabouts history, testing frequency, the ten-year sample storage window and retrospective medal reallocation — none of it was available. The 40mm sole cap, the DSD rules, individual entry as an ANA: none of it could be attached to a person. The sixth is the transmission chain. Athletics monetises on the Olympic cycle, and a single news item's value depends on where in that cycle it sits. Time sensitivity was never assessed, so the position is unknown. And when the position is unknown, the real Indian question stalls exactly where it starts: sports-quota jobs at Railways, Services and the PSUs, TOPS funding, state cash awards — a structure that runs on the calendar, not on form. The seventh is the cost of re-checking. I rebuilt the whole audit from scratch, then checked it twice. The result held. Three seasons of events, one question — who is actually doing the work? — and an empty file for an answer. Then the part where I had to fight my own reflex. The first urge was to hunt a discrepancy, something that would prove everyone wrong. ESTJ certainty gets restless on occasions like this. But there was no discrepancy to find. When the accepted number survives scrutiny, letting it survive is itself a finding, and it deserves a paragraph. The second habit is more dangerous — turning a correction into combat. The easy route is to call a reporter or an editor careless. Nothing changes that way. The real cause is structural: no dedicated beat, no storage archive, dependence on agency copy, result PDFs arriving months late. When the system has no archive, every reporter starts from the previous reporter's copy, and the error becomes inherited property. From the 0.01 seconds at Los Angeles 2026 to the mis-attributed 2026 silvers, Indian athletics' record book is full of exactly these inheritances. The third habit is reading N/A as 'no risk'. That is the single biggest risk in this whole report. A reader who sees N/A nine times will conclude, in a hurry, that everything is clean. The truth is the reverse: N/A means the check was never run, and an unrun check is never a clean chit. What to watch next comes down to three signals. One: if the original document is re-ingested, at least a title and one information point should return — if they do, the fault sat in the extraction layer, not the source. Two: compare information-point count against article length; zero points from a full document is a bug, zero from an empty one is not. Three: backfill the publication date — without it there is no heat cycle, no peaking window, no accountability. For now, file this one away — but do not erase the N/A. The day an empty column gets read as a zero is the day the next error has a clear road.