Six-Minute Man Sander Eitrem 5:58.52 -- first man below the 6:00.00 barrier

TL;DR

The paper applies extreme-value theory to the 5,000 m speedskating sub-6:10 records up to the 2024–2025 season, fitting a two-parameter model with $G(y,a,σ)$ and $g(y,a,σ)$ to predict the likelihood of breaching the 6:00 barrier and achieving a world record. A Poisson framework is used to translate season-level extremes into probabilities for event timings, yielding a pre-season world-record probability of $p({\rm wr})\approx0.109$ and a $6:00$-barrier probability of $p(6:00)\approx0.012$, both with skewed confidence curves. The analysis introduces a monitoring process $Z_n(y)$ to assess stationarity, finds no significant drift through 2024 and then notes regime changes in 2025–2026 that alter predictions (e.g., post-Eitrem $\hat γ\approx18.74$, $r_0\approx5:51.26$). It further interprets the ultimate speed bound via $r_0=6:10.00-\gamma$ with $\gamma=\sigma/a$, discussing an extreme but plausible ‘Ultima Thule’ 5k scenario and highlighting future extensions to other distances and sports. Overall, the work provides a principled probabilistic framework for forecasting record-breaking performances in speedskating using extreme-value theory and Poisson=Nonebased event modeling, while acknowledging data-driven regime shifts and their impact on long-horizon predictions.

Abstract

In Calgary, November 2005, Chad Hedrick was the first to skate the 5,000 m below 6:10. His world record time 6:09.68 was then beaten a week later, in Salt Lake City, by Sven Kramer's 6:08.78. Further top races and world records followed over the ensuing seasons; up to and including the 2024-2025 season, a total of 126 races have been below 6:10, with Nils van der Poel's 2021 world record being 6:01.56. The appropriately hyped-up canonical question for the friends and followers and aficionados of speedskating has then been when (and by whom we for the first time would witness a below 6:00.00 race. In this note I first use extreme value statistics modelling to assess the state of affairs, as per the end of the 2024-2025 season, with predictions and probabilities for the 2025-2026 season. Under natural modelling assumptions the probability of seeing a new world record during this new season is shown to be about ten percent. We were indeed excited but in reality merely modestly surprised that a race better than van der Poel's record was clocked, by Timothy Loubineaud, in Salt Lake City, November 14, 2025. But Six-Minute Man Sander Eitrem's outstanding 5:58.52 in Inzell, on January 24, 2026, is truly beamonesquely shocking. I also use the modelling machinery to analyse the post-Eitrem situation, and suggest answers to the question of how fast the 5,000 m ever can be skated.

Figures (4)

• Figure 1: There has of course been room for S. Eitrem's Overtro og trolldom hos de gramle grekere on my bookshelves (from 1921, and with a new translation of Lukan's Lø gnhalsen), since S. Eitrem first showed world-class 5k promise in 2021, i.e. hundred years later.
• Figure 2: Left panel: all 126 sub-6:10 races, from 2005 to 2025, where speedskating enthusiasts easily can spot and identify world records by Hedrick, Fabris, Kramer, Bloemen, van der Poel. Also included are the so far surprisingly high number 39 of such races for the 2025-2026 season, pre the Milano Olympics, with Eitrem's beamonesque 5:58.52. Right panel: empirical and fitted c.d.f.s for sub-6:10 races, up to 2024-2025, with van der Poel's world record 6:01.56.
• Figure 3: Left panel: the probability curve, for seeing a race at a certain level, or better, during the 2025-2026 season, from the pre-season perspective, using $\lambda=25$ in the (\ref{['eq:withpoisson']}) formula. Ths probabilities for seeing a new world record, or a six-minute race, are estimated at 0.109 and 0.012. Right panel: confidence curves for the two probabilities. These are non-tight and quite skewed, so quoting the 10.9 percent chance of seeing a world record is telling merely half the story.
• Figure 4: Left panel: the model monitoring process $Z_n$ based on all $n=126$ races from 2005-2006 to 2024-2025 (full black line), along with ${\rm sim}=25$ simulated such processes (thin red lines), under model conditions. Right panel: the Ultimate Race time, $r_0=6\colon\!\!10.00-\widehat{\gamma}$, estimated as 5:57.46 with pre-season data (red dashed) but now post Eitrem as 5:51.26 (full black), with confidence curves.