Generations
Who races it, by age
520,575 classified finishes across 12 editions, placed in the age band the organiser filed each athlete under. A further 6 carry no usable age class.
French athletics, in two masters schemes — five-year M-bands in the recent editions, ten-year V-bands in the older ones.
Classified finishes, all editions. Hover a bar for its count.
Each band’s share of the classified field that year, young at the bottom. Hover a year for its make-up.
The bands sum to the whole classified field each year, so this is composition — it says nothing about whether the race grew.
Is the mapping right?
The organiser’s class codes do not state their ages, so they are mapped — and the mapping is checked against this archive rather than taken on trust. If it were off by one class, anyone racing either side of a vocabulary change would appear to jump years in a single season.
Across 91,529 pairs of consecutive editions by the same person, the band moves backwards 2,105 times and skips more than one bucket 569 times — 97.1% consistent. The residue is the size name collisions predict.
Median finish by band
| Band | Marathon de Paris |
|---|---|
| Under 20 | 5:16:15 · 17 |
| 20–39 | 4:36:22 · 262,012 |
| 40–49 | 4:33:13 · 153,700 |
| 50–59 | 4:48:50 · 84,813 |
| 60–69 | 5:14:40 · 18,243 |
| 70 and over | 5:52:20 · 1,790 |
Each cell is the median of that band’s per-edition medians, with the classified finishes behind it.
About this project and its data. The results are Sport Innovation / TimeTo’s — the official timekeeper of the event. They were fetched once after the event and republished here with segment times, recomputed leaderboards and a page per athlete that the official results do not offer. The timing is their work; the analysis is mine, and any mistake in it is mine too. Athletes who asked to be anonymous in the official results stay anonymous here.
Computed from public_dashboard.race_* over all 12 editions · built 2026-09-08