Nicolas Sitter
The La Marmotte archive

Regulars

Who keeps coming back

21,685 distinct people have finished at least one of the 7 editions. 19,280 came once and did not return; 2,405 came back at least once more.

10 people have finished every one of the 7.

How many editions people finish

Each bar is the number of people with that many finished editions. Hover for the count.

05,00010,00015,00020,0001234567

The first step is the steep one — 89% of people who finish never come back. The right-hand tail is real but invisible at this scale; it is in the table below.

4 editions or more — 32 people

“Editions” counts distinct years, not finishes.

NameEditionsFinishesYearsBest
Tim ALLEMAN7720 21 22 23 24 25 265:23:28
Stijn LAUREYS7720 21 22 23 24 25 266:12:18
Benoit ARSENNE7720 21 22 23 24 25 266:55:00
Jim PETERS7720 21 22 23 24 25 266:58:43
Thomas BIEBERSTEDT7720 21 22 23 24 25 267:01:08
Tom SAXLUND7720 21 22 23 24 25 267:06:50
Dirk OST7720 21 22 23 24 25 267:29:05
Leon RENET7720 21 22 23 24 25 267:44:43
Pierre MIRB7720 21 22 23 24 25 267:51:27
Craig NORTHOVER7720 21 22 23 24 25 268:17:42
Fabio LANZONE6621 22 23 24 25 265:25:12
David POLVERONI6620 22 23 24 25 265:35:03
Gabriel BOIS6621 22 23 24 25 266:32:02
Roger MOVERLEY6621 22 23 24 25 266:44:17
Pierre OURNIER6620 21 22 23 24 266:47:34
Yannick COLLOT6620 21 22 24 25 266:48:56
Bart HAVERMANS6620 21 22 23 24 266:53:26
Esteve SERRAT GENESCÀ6621 22 23 24 25 267:05:05
Gregory DELMEE6620 21 22 23 24 257:06:17
Koen DEROUS6621 22 23 24 25 267:07:27
Samuel PEPINO6620 22 23 24 25 267:08:48
Victor BOLWERK6620 21 23 24 25 267:14:31
Andre SCHIEBEL6621 22 23 24 25 267:27:22
Pascal CHARRE6620 21 23 24 25 267:31:10
Thomas VERYEPE6620 21 22 23 24 257:35:30
Gavin LEWIS6620 22 23 24 25 267:50:25
Antoine ARNOUX6620 21 22 23 24 268:02:29
Alexandre MOZINA6621 22 23 24 25 268:04:42
Laurent MAILLET6620 21 22 23 25 268:30:27
Michael HANSEN6620 21 22 23 25 268:31:47
Mario Davide RUSSO6620 21 22 24 25 269:01:03
Martin PORTER6621 22 23 24 25 269:57:55
How sure is “the same person”? These exports carry no athlete id, so one person across editions is a normalized name — accent- and case-folded, punctuation stripped, first and last name treated as interchangeable because the organiser swaps them between years. That merges different people who share a name, and the rate is measured rather than assumed: two finishers in the same race are certainly different humans, and 22 such collisions occur among 25,054 named finishes. Applied to every cross-edition pair that predicts roughly 117 merges that are two people rather than one. Read the repeat counts as an upper bound.

About this project and its data. The results are ChronoRace’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 7 editions · built 2026-09-08