Tour Divide: a comparison of Justinas Leveika’s FKT and Robin Gemperle’s performance 2025

Robin Gemperle's victory in the 2025 Tour Divide is historic in two ways: never before has a rider dominated the competition as Robin did, and never before has a rider finished as fast as Robin. And yet, it remained unclear until the very end whether Robin would be able to claim the Fastest Known Time (FKT) because he was forced to take a detour due to forest fires (which was longer than the original route). 

An anonymous dotwatcher took the trouble to compare the races of Justinas (who set the record in 2024) and Robin (2025). The result was that Justinas would probably not have had a chance against Robin. Below is the anonymous dotwatcher's analysis in its original form.

TL;DR: A detailed comparison of Robin’s (2025) and Justinas’ (2024) times reveals that Robin was generally faster, but differing conditions (weather, competition) and strategic choices (sleep, breaks) played a crucial role. Both performances are exceptional and highlight the ongoing evolution of bikepacking. Or perhaps the last two years represent a revolution, fueled by the incredible rides of Lachlan, Uba, and these two competitors.

Above all, health is paramount, and I wish all riders – those at the front, in the midpack, at the back, and especially those who unfortunately had to withdraw – all the best and a swift, full recovery. I also extend my best wishes to all those deeply impacted by the forest fires.

I’ve compared the performances of Justinas (2024) and Robin (2025). It’s important to acknowledge that a direct comparison of two races is inherently difficult, even within the same year. Factors such as snow and rain in the north, heat and wildfires in the south, the infamous “Peanut Butter Mud,” wind, competition, personal fitness, illness, and unforeseen events all contribute to varying race dynamics. Nevertheless, I undertook this comparison to gain a deeper understanding of their respective journeys.

The data originates from Trackleaders.com (here and here). The graphic is based on the 21 checkpoints defined by Trackleaders.com, supplemented by additional checkpoints I manually identified. These additions primarily mark locations where riders took extended breaks, such as for sleep. I identified a total of 46 segments, ranging in length from 14 to 143 miles. I meticulously recorded the arrival and departure times of each rider at each checkpoint.

This analysis shows that Robin averaged approximately 62 hours of sleep over 12 nights, while Justinas managed around 46.5 hours of sleep over 13 nights. Justinas frequently utilized short roadside naps, whereas Robin, with the exception of the first afternoon and the penultimate night, prioritized longer, more restorative sleep periods in hotels. It’s important to note that these sleep times encompass not only rest but also time spent on essential tasks like check-in, shopping, showering, bike maintenance, and similar activities. I manually extracted these times from the Trackleaders data. The automatically calculated stopped times on the platform are higher (approximately 12 hours for Robin, almost 24 hours for Justinas) because they include brief pauses for everyday needs – obtaining water, applying chain oil, quick recovery moments, and interacting with DotWatchers.

Robin demonstrated remarkable efficiency in these tasks, benefiting from generally more favorable weather conditions. For the approximately 2700 miles, Robin’s total riding time (including short stops) was around 221:15 hours, compared to Justinas’ 268 hours.

At each waypoint, I calculated Robin’s lead upon arrival and departure, and used this data to create the attached graphic. In 5 segments, Justinas was faster, consistently occurring shortly after a sleep break for him and just before one for Robin – notably once in the altered segment before Steamboat Springs. In all other segments, Robin was able to build his lead, particularly on the Gila Detour necessitated by wildfires, where he gained approximately 2 hours. Moreover, Robin gained significant time over Richmond Peak, further north around Lava Mountain and Fleecer Ridge, throughout Idaho, in the Great Divide Basin, between Sargents and Del Norte, and in the particularly challenging sections of New Mexico.

Finally, several questions remain: Was Robin’s speed primarily attributable to the weather? Was he faster due to the reduced competition in 2025, allowing him to execute his (sleep) plan undisturbed? Or was he, perhaps, less driven by the lack of an external push, unlike Justinas, who was spurred on by the weather and his friend and rival, Uba?

It is important to note that this analysis does not take a position on which of these times should be considered the official Fastest Known Time (FKT). As I believe both were tremendous accomplishments, it is so unfortunate that Justinas had bad luck with the weather up north, while Robin had bad luck with the fire detour.