Welcome to The Extra Lap! Each week, we’ll be sharing HYROX insights and analysis, product updates, stories, community shoutouts and much more.

Subscribe here!

Stat of the Week

While HYROX London had the most sub 60 men’s pro finishers (36) in an event all season, London only ranked 7th for men’s pro sub 60 finishers per 100 finishers (3.8 finishers).

Stuttgart is #1 this season with 7.1 sub 60 finishers per 100 (24 total).

For women, London also had the most Sub 65 finishers all season (15) but ranked 11th for sub 65 finishers per 100 finishers (2.5 finishers).

Stuttgart is also #1 with 5.8 finishers per 100 (9 total).

Are Sleds More Consistent?

As many of you know, HYROX implemented a new standardized sled turf from Centr this season to improve consistency.

Fast vs slow sleds has always been a never ending debate and we commend HYROX for investing in new turf and prioritizing the issue vs just ignoring the feedback.

However, now with half the season almost complete, we wanted to analyze whether this new turf has indeed brought more consistency.

Now, we think about consistency from two angles: venue-to-venue and within the same venue.

Let’s dive into how we tackled this.

We filtered for pro division race results from the current season (June - November 2025) and the same comparative period last season (June 2024 - November 2024).

If the new Centr turf has brought more consistency, we would expect a smaller relative variability venue-to-venue this season.

Venue-to-Venue Analysis

I’ll spare the technical details, but for those interested, we took the median of each course for men and women, calculated the standard deviation across each season comparative period, and then calculated the coefficient of variation (“relative variability”).

Here are the results.

Interestingly, venue-to-venue, sled pull performance is more consistent than last season but sled push is more volatile.

However, we wonder if the added sled push volatility may just reflect the new surface being less forgiving unless athletes get low and push correctly, something not everyone has adjusted to yet which doesn’t apply to sled pull.

Within Venue Analysis

For consistency within the same venue, obviously we don’t know which lane an athlete used.

However, can take a similar approach and measure the relative variability within each course opposed to course-to-course variability above.

And that’s what we did. We calculated the relative variability (coefficient of variation) for men and women pro athletes for each course. We then took the average for both seasons.

Here we see a similar story. Sled push is more variable whereas sled pull improved, though women’s is actually slightly worse.

London Bounty Didn’t Bankrupt Us

Whew! It’s definitely a little scary offering large cash prizes for beating our model but at the same time we have a lot of confidence, having battle tested and refined our system for 2+ years.

Our peak (excellent) prediction is our personalized estimate of your fastest possible time in your next race. Technically, under the hood, the model is saying, “We think there is a 5% chance this time will be beat.” If we are accurate, it will only get beat 5% of the time each week.

Low and behold, despite the 1 lap London course and some crazy fast times, still, only 5% of athletes beat the RoxOpt peak this weekend.

Thankfully, with courses like Berlin and New York, we can account for course uncertainty in historical data and still maintain solid accuracy with outlier courses.

At the end of the day, we are trying to predict what’s possible. And when it comes to HYROX, the possibility of a fast vs slow course is relevant.

Because of this, we LOVE when people beat it. Doing so means they truly had an incredible performance and that is worth celebrating.

As a matter of fact, Cole Learn, the first customer to collect a bounty this season back at HYROX Boston, is competing in his first Elite 15 Major this week in Melbourne.

RoxOpt Origin Story - Entry #6

Again, it’s June 2023 and I was on a call with Rich Ryan for the first time, picking his brain.

Before I started building RoxOpt, I assumed integrating wearable data would be priority number one.

However, the most surprising feedback I got from Rich was that that didn’t get him as excited or solve a burning pain point he had.

What he said really stuck with me.

Accuracy can vary device to device. While it’s a helpful training tool, during a race it can sometimes be a detriment if one fixates too much on it. An athlete may be feeling great but could see a heart rate spike and suddenly get way too in his/her head.

Wearable data is certainly relevant but it isn’t the end all be all in terms of what could be helpful for HYROX.

This was very valuable feedback and helped me avoid over-complicating the first iteration of RoxOpt… though I will say we have exciting wearable features on our roadmap for 2026.