Six Years, Two Charges: Ruti Aga's Career Was Ended by a Prescription Pad, Not an Injection

September 10 appears twice in the Athletics Integrity Unit's Thursday statement: once as the start of Ruti Aga's provisional suspension, once as the anchor point of a six-year ban. No other number in the file carries the same weight. The half marathon she won in China in March has no published time, no splits, no field depth, no weather, no course profile. There is no data here. That is the finding.

Aga is 32, Ethiopian, the 2026 Tokyo Marathon champion, three times a runner-up at a World Marathon Major. When the ban expires she will be roughly 38. In Ethiopian women's marathoning, a 38-year-old returning from a two-year absence with a doping record on file does not get selected. That is not a rule; it is selection arithmetic.

The facts that do exist are three: a positive sample for methylprednisolone, a cited knee injury, and a prescription the AIU has ruled forged. On the third point most outlets made their first error, calling methylprednisolone a prohibited steroid. It is a steroid, but not an anabolic one. It is a glucocorticoid, WADA Prohibited List class S9, prohibited in the current List in-competition only, and only by injection, oral or rectal route. It does not carry the year-round prohibition that attaches to class S1 anabolic agents.

That one technical correction reorganises the case. The dispute was never about the molecule; it was about route and timing. A therapeutic knee injection given out-of-competition is entirely lawful. It becomes a violation only when residue is present in an in-competition sample, and a road race's testing window opens at 23:59 the day before the start. An injection three days out gets caught; one five days out may not. The whole file lives inside that window.

Now the economy, because that is where the file comes from. Between the Majors and the domestic scene sits a middle market for East African marathoners: Asian label races, appearance fees, thin fields, low risk, guaranteed money. For a 2:18 to 2:20 runner, that tier is where the living fee is paid, because the Major medal fee is not. That same tier is now the AIU's most productive testing ground. The appearance-fee economy and the testing economy are converging on the same start lines: the race that signs the cheque also collects the sample.

What is missing from that structure is a credentialed team physician. Per the AIU statement, the prescription was written by a family member with no medical registration, who claimed to have used “an old prescription pad from a former workplace.” The clinic named on the document said it had never issued it. That is not a filing error. It is the signature of an athlete operating with no formal medical support at all.

I have spent years treating unpublished documents as a subject rather than a gap. In 2026, furloughed to three days a week, I wrote a scraper against AFI and state-meet result PDFs, 1,400 documents spanning 2026 to 2026, and found published 400m split data for exactly three Indian meets in a decade. What is not published does not practically exist, and what does not exist cannot be audited. This case is the other terminus of the same logic: a document existed, and it was false.

The knee deserves to be stated before the arithmetic. Chronic knee synovitis is ordinary in high-mileage marathoners, and the standard sports-medicine treatment is exactly this corticosteroid injection. The incentive to mask pain is also real, because appearance fees do not survive a medical withdrawal. For a runner without a team doctor, this drug is the cheapest available tool for both problems. The forgery is downstream of that gap, not the cause of it.

Now the ledger. Two years plus four years equals six. Two years for the presence violation, meaning a prohibited substance in the sample. Four years for tampering, meaning a fabricated document submitted and a misleading account given under interview. The addition matches the standard Code architecture, and the arithmetic is where the information sits.

Two years tells us the panel classed the substance as a Specified Substance and did not find proven intent; the default would otherwise have been four. Four years tells us the panel treated the forgery as fully culpable and rejected the no-significant-fault mitigation. The six is not built from the substance. It is built from the paper.

There is a paradox inside that structure. A forged prescription is itself circumstantial evidence of concealment intent; it nonetheless did not aggravate the presence sanction. The panel treated the two as independent violations rather than multipliers of each other. That is technically clean and judicially arguable.

Backdating to 10 September is not relief. Time served under provisional suspension counts against the ban, so the athlete gains nothing in real terms. The clock was already running from the day she could not race.

The question the statement does not answer, and the one that matters more to her record than the six years, is the scope of disqualification. Standard AIU practice voids the competition at which the sample was collected, meaning the Chinese half marathon, its title and its prize money; ranking points accrued after that date go with it. The 2026 Tokyo title does not fall automatically, absent a separate finding of earlier use. The report simply does not state the scope.

The 2026 title is her actual asset — three Major silvers, one Major gold, a best in the 2:18 range. A six-year ban removes her from racing. A disqualification scope removes her from the record. Those are different punishments, and the coverage has collapsed them into one.

My own method applies here. In 2026, building an automated match-report engine in Chicago, my first pitch was that a 2-1 win was really a 0.4-2.3 xG defeat. The desk editor spiked it; I rewrote it with the shot map as the lede and it cleared. The lesson was that the number is the story, not the footnote. Here the number is not a scoreline but a subtraction of dates: 2 plus 4, and 32 plus 6.

On the Russia World Cup live desk in 2026 I learned the companion lesson — a correct number published on time beats a beautiful number published four hours later. Forty minutes after Germany lost 0-2 to South Korea in Kazan we ran the pressing-collapse chart: German PPDA had drifted from 8.1 at Brazil 2026 to 13.6 across the 2026 group stage. The same discipline applies to a disciplinary file. Print what the statement says; record what it omits — the year of the race, the mark, the disqualification scope — as an absence, not a guess.

The third failure in the coverage concerns the nature of the evidence. The four years did not come from a document alone; they came from an interview, meaning statements made under questioning. Language and translation conditions in AIU interviews with Ethiopian athletes are a recurring friction point. That does not reduce the sanction, but it identifies what the case actually rested on: not the paper, the mouth.

Here is where the easy conclusion goes wrong. The easy conclusion is that this is a doping case, the punishment is severe, and the system is working. All three are half true.

This is not a doping case in the sense the reader assumes. A glucocorticoid has no proven endurance benefit; it is an anti-inflammatory. Outlets calling it a steroid are inviting a comparison with cases it should not be compared to. A category error at the classification stage corrupts the benchmark for severity.

The second easy conclusion is that this is one athlete's ethical lapse and therefore no system question arises. Read it the other way. An athlete with a credentialed team physician either holds a TUE or picks a date outside the competition window. Here three independent verifications failed at once — no clinic, no registered doctor, a pad the clinic disowned. Any single one of them holding would have prevented the file from reaching this size.

The third, and the most delicate: one file does not map a system. I have no dataset on how many other athletes in the same training cluster are running on informal treatment of this kind. The pattern that appears is recognisable — East African road-racing files where the initial substance violation is compounded by a fabricated explanation. But a pattern that looks like a pattern is not a proven pattern. Correlation is not causation. I went looking for a trend and found a hole; naming the hole is more honest than filling it with inference.

What remains is proportionality. Six years for a substance with no proven endurance benefit will be cited both as evidence of AIU rigour and as evidence of disproportionate enforcement against African road runners. Both claims are unverifiable until the reasoned decision is public. I prefer to show the receipts before the conclusion, and this receipt has not arrived.

Three signals to watch. First, the appeal window — typically around 21 days from the reasoned decision at CAS; a document-based finding is hard to overturn, but an appeal can reach part of the six. Second, the Ethiopian federation's silence: no national-level statement appears in the statement, and that silence is itself a data point. Third, testing pressure on China's label-race circuit, where appearance-fee budgets and sample collection now share the same stadium.

The larger signal is that the enforcement frontier has moved. This case was decided by document verification rather than by detection. In the sport with the thinnest published data, the next frontier is verifying the paperwork. The next time a road-race result and cheque are voided, watch whether the coverage fixes on the six years or on the single forged pad that produced them. The question is not whether the sanction was correct. It is that the sanction's strongest evidence was the absence of a clinic, and nobody saw that absence first.