On 15 February 2026, Mohammed El Yamani ran 2:28:28 at the Seville Marathon. He was 61.
Run that through the RaceRecords engine and it scores 100.76%. That's the first marathon I've found that clears 100% on the current Alan Jones 2025 tables. Ten weeks later Sabastian Sawe ran 1:59:30 in London and scored 100.91%. Two more men in the same race cleared it too.
In June I wrote that none of the ten greatest masters marathons reach 100%, and that this was the marathon being the marathon. It was already wrong. El Yamani had run his 100.76% four months earlier, and I missed it. So this is the correction, and an attempt to explain why 100% isn't the wall it looks like.
Did a marathon really age-grade over 100%?
Yes, four of them in 2026, all men. On the Alan Jones 2025 road table, El Yamani's 2:28:28 at 61 scores 100.76%. Sawe's 1:59:30 scores 100.91%, Yomif Kejelcha's 1:59:41 scores 100.75% and Jacob Kiplimo's 2:00:28 scores 100.10%. The last three came from the same London race on 26 April (World Athletics ratification release, 2026-07-23).
Before February, the highest masters marathon age grade I could verify on these tables was Tommy Hughes at 99.15%, for 2:27:52 at 59 in Frankfurt 2019. The only run that had ever sat at exactly 100% was the record the standard was built on.
| Runner | Age | Time | Race | AG% |
|---|---|---|---|---|
| Sabastian Sawe | open | 1:59:30 | London 2026 | 100.91% |
| Mohammed El Yamani | 61 | 2:28:28 | Seville 2026 | 100.76% |
| Yomif Kejelcha | open | 1:59:41 | London 2026 | 100.75% |
| Jacob Kiplimo | open | 2:00:28 | London 2026 | 100.10% |
| Tommy Hughes | 59 | 2:27:52 | Frankfurt 2019 | 99.15% |
The three London runners are all in the open age range, where the factor is 1.0000. So their score is simply 2:00:35 divided by their time. Nothing about age is involved. El Yamani is the interesting one.
How did a 61-year-old run 100.76%?
Mohammed El Yamani ran faster than the Alan Jones 2025 tables think a 61-year-old can. The M61 marathon factor is 0.8061, which puts the age standard at about 2:29:35. El Yamani ran 2:28:28, a minute and seven seconds under it.
So the arithmetic. 2:28:28 is 8,908 seconds. Multiply by 0.8061 and you get an open-class equivalent of 7,181 seconds, or 1:59:41. Divide the 2:00:35 standard by that and you get 100.76%.
That open equivalent is the same as Kejelcha's actual time in London, to the second. In age-graded terms, a 61-year-old tied the second-fastest marathon in history.
For context, the previous M60 best was Hughes's 2:30:02 in Lisburn in October 2020. El Yamani took 1:34 off it (Wikipedia, M60 marathon record progression, retrieved 2026-10-07). World Athletics gives his date of birth as 4 September 1964, so he was 61 years and 164 days old on race day.
Why do the tables still use 2:00:35?
The Alan Jones 2025 tables still use 2:00:35 because they're a fixed edition, not a live feed. The Alan Jones 2025 road standards were approved by USATF's Masters Long Distance Running Council on 10 January 2025 (Alan Jones, 2025 Readme). At that point the men's record was Kelvin Kiptum's 2:00:35 from Chicago on 8 October 2023, and that became the open standard.
The record book updates the day a record is ratified. The tables update when someone publishes a new edition. The road editions so far are 2015, 2020 and 2025, so I wouldn't expect the next one soon. Until then, anyone who beats 2:00:35 grades over 100%, and every masters runner keeps being measured against Kiptum rather than Sawe.
The women's standard isn't affected. It's 2:09:56, Ruth Chepng'etich's time from Chicago in October 2024, and it's still the women's record (Wikipedia, marathon world record progression, retrieved 2026-10-07).
When the standard moves to 1:59:30
Every men's marathon age grade drops by about 0.9% of itself. That's the ratio of the two standards, 7,170 seconds over 7,235. If a future edition adopts Sawe's time and keeps the same age factors, El Yamani goes from 100.76% to 99.85%. Sawe lands on exactly 100.00%, which is the point of a standard.
| Runner | Now (2:00:35) | Against 1:59:30 |
|---|---|---|
| Sabastian Sawe | 100.91% | 100.00% |
| Mohammed El Yamani, 61 | 100.76% | 99.85% |
| Yomif Kejelcha | 100.75% | 99.85% |
| Jacob Kiplimo | 100.10% | 99.20% |
| Tommy Hughes, 59 | 99.15% | 98.26% |
| Ed Whitlock, 80 | 98.98% | 98.10% |
That's an assumption, so read the right-hand column as a sketch. A new edition doesn't just swap the open standard. The age curve is refitted to the best performances at each age, and El Yamani's run is now the best at 61. A refit that pulls the M61 standard toward his time would take his score lower still. Either way, his 100.76% is a feature of this edition, not a permanent number.
I've seen this happen to my own result before. When the record moves, my NYC Marathon age grade moved with it. The run didn't change. The ceiling did.
What 100% actually means
100% means you matched the model's world standard for your age and sex. It doesn't mean you matched the world record, and it isn't a limit. The tables are a curve fitted through the best times at each age, so a few real performances can sit above it.
That's why I keep saying age grading compares you to the world best, adjusted for age, rather than "the world record for your age". Clare Elms's W60 1500m world record graded 98.9%, below the curve. El Yamani's M60 age-group record grades 100.76%, above it. Both are records. The curve sits in different places relative to each one.
The marathon used to look like it couldn't get above 100% at all. I put that down to the distance. A better explanation is simpler: until 2026, nobody had run the time.
Does this change my own age grade?
Not today. RaceRecords grades every road race on Alan Jones 2025, and that edition hasn't changed, so your marathon age grade is the same as it was in March. The methodology page lists the tables in use.
When the next edition does land, expect a drop of about 0.9% of your score, plus whatever the refit does at your age. For a 2:59 at 50 that's 75.16% now and 74.48% against 1:59:30, about 0.7 of a point. For a 3:00 at 40 it's 68.48% to 67.86%. Smaller than a bad day, but it'll happen to every man's marathon at once, so it's worth knowing in advance.
If your number moves and your time didn't, here's why calculators disagree. And if you want to know where your own marathon lands against 2:00:35, what your age-graded percentage means is the place to start.
Sources
Age grades computed with the RaceRecords engine on the Alan Jones 2025 road table (open standards 2:00:35 men, 2:09:56 women). The "against 1:59:30" column is 7,170 / (time × factor) with the published factors unchanged.
- World Athletics, Zurich Maratón de Sevilla 2026 results. El Yamani 2:28:28, 15 February 2026, date of birth 4 September 1964. Retrieved 2026-10-07, https://worldathletics.org/competition/calendar-results/results/7236004
- World Athletics, ratified world records press release, 23 July 2026. Sawe 1:59:30, Kejelcha 1:59:41, Kiplimo 2:00:28. Retrieved 2026-10-07, https://worldathletics.org/news/press-releases/ratified-world-records-sawe-ehammer-charlton-mcrae-yan-duplantis-assefa
- Wikipedia, "Masters M60 marathon world record progression". El Yamani 61 years 164 days, Hughes 2:30:02 Lisburn 2020. Retrieved 2026-10-07, https://en.wikipedia.org/wiki/Masters_M60_marathon_world_record_progression
- Wikipedia, "Marathon world record progression". Kiptum 2:00:35 Chicago 2023, Chepng'etich 2:09:56 Chicago 2024. Retrieved 2026-10-07, https://en.wikipedia.org/wiki/Marathon_world_record_progression
- Alan Jones, Age-Grade Tables, 2025 Readme. Approval by the USATF MLDR Council on 10 January 2025. Retrieved 2026-10-07, https://github.com/AlanLyttonJones/Age-Grade-Tables/blob/master/2025%20Files/Readme.txt
- Alan Jones, Age-Grade Tables, 2025 road factors (M61 marathon 0.8061). Retrieved 2026-10-07, https://github.com/AlanLyttonJones/Age-Grade-Tables