HYROX Running Study: What 39,696 PRO & ELITE Results Really Mean

Running Wins the Clock — But It Does Not Win HYROX Alone

What 39,696 PRO and ELITE results mean for your running, strength, and race strategy.

A new analysis of 39,696 PRO and ELITE race results gives HYROX its clearest performance signal yet. Running matters enormously but the strongest coaching lesson is more precise than “just run more.”

HYROX is built on a simple rhythm: run, work, repeat. Eight 1-km runs are broken up by eight stations, creating a race that demands aerobic power, threshold speed, strength, technical efficiency, pacing, and fatigue resistance.

But when nearly 40,000 PRO and ELITE performances were analyzed across the sport’s first seven seasons, one signal stood above the rest of the clock: running.

The study, published July 8, 2026 in Frontiers in Physiology, found that running consistently occupied about half of total race time. Running rank also aligned more closely with overall finishing rank than any single station, and faster running accounted for the largest share of the improvement seen in Top-100 performances over time (Rappelt et al., 2026).

Running is the broadest performance lever in HYROX. The complete lesson: build running capacity, protect it with an adequate strength floor, and train the ability to reproduce pace after stations.

How the researchers studied HYROX performance

This was not a laboratory experiment or a training intervention. It was a retrospective longitudinal analysis of publicly available race results from the official HYROX database. The researchers extracted individual PRO and ELITE results from seven seasons, spanning 2018/2019 through 2024/2025.

The starting dataset contained 51,469 records. After duplicate listings were removed, the authors excluded performances with penalties, time bonuses, or missing discipline-specific splits. The final analysis included 39,696 results: 27,854 male and 11,842 female performances.

A result is not necessarily a unique athlete. Some competitors raced more than once. The researchers addressed that concern with sensitivity analyses that retained one performance per athlete per season and accounted for event-level clustering.

The study used several complementary approaches:

• Percentile curves to examine how the wider PRO field changed across seasons.
• Top-100 comparisons to study high-performance changes in running, stations, Rox Zone time, and total race time.
• Rank reshuffling to test how closely each discipline’s ranking resembled final race ranking.
• Quantile regression to evaluate discipline associations across faster and slower performers while removing each discipline’s own seconds from total time.
• ELITE medians and coefficients of variation to determine whether the highest-level fields became faster and more competitive.

WHY THE DESIGN MATTERS Because this was observational race-result research, it can reveal strong patterns and associations. It cannot prove that a particular increase in running volume or a specific training method caused athletes to become faster.

Finding 1: Running owned about half of the clock

Across seasons and sexes, running represented approximately 50% of total race time. That is the largest direct time footprint in HYROX. No station comes close to occupying the same number of minutes.

This sounds obvious — HYROX includes 8 km of running — but the size of the dataset makes the practical implication much harder to ignore. A small percentage improvement across eight runs can produce more total time than an aggressive gain on one short station.

That does not mean every athlete should immediately increase run mileage. It means coaches should treat running as a primary system within the program, not as filler between functional exercises.

Finding 2: Most Top-100 improvement came from running

The Top-100 male mean improved from 70:13 in Season 1 to 56:56 in Season 7 — a total gain of 13:17. Cumulative running time improved from 35:44 to 27:38, accounting for 8:06, or approximately 61%, of the total improvement.

The Top-100 female mean improved from 80:43 to 63:11 — a gain of 17:32. Running improved from 40:20 to 30:17, a 10:03 change representing approximately 57% of the total improvement.

Coaching Tips:The faster field did not improve only by attacking machines and stations. A large share of the gain came from protecting more speed across all eight runs.

Finding 3: Running rank most closely resembled finishing rank

The authors compared each athlete’s discipline rank with the athlete’s final rank. Running consistently produced the lowest rank reshuffling. In Season 7, the normalized rank-difference standard deviation was 12.7% for men and 12.5% for women — lower than every station.

Thus, where an athlete ranked as a runner tended to resemble where that athlete finished overall. Running acted as the most stable ranking signal across the field.

However, low reshuffling does not mean stations were unimportant. Stations changed positions more dramatically, particularly when an athlete lacked the strength, skill, or local muscular endurance needed to prevent a major time loss.

The important nuance: the headline is true — but incomplete

Running had the largest direct contribution to total race time and the closest alignment with final rank. But the study’s adjusted models asked a different question: when the discipline’s own seconds were removed from total race time, how strongly was that discipline associated with performance during the rest of the race?

In those models, running had intermediate independent explanatory value. Rowing produced the highest pseudo-R² range in men, while rowing, sandbag lunges, burpee broad jumps, and several other disciplines sometimes matched or exceeded running in women.

This is not a contradiction. Running directly controls more of the clock because it lasts longer. Rowing, lunges, burpees, and other tasks may act as markers of broader aerobic durability, technical proficiency, muscular endurance, pacing, and fatigue resistance.

Running controls the broad clock. Stations expose the athlete’s missing link — and can create disproportionate damage when that link is weak.

Why slower athletes lose more time at stations

The quantile-regression coefficients generally became larger among slower performers. In practical terms, weak station performance created a larger absolute penalty in the slower parts of the field, especially during strength-oriented stations.

A sled that is merely slower for a prepared athlete can become a survival event for an athlete without sufficient force reserve. The cost is not limited to the station split. Excessive local fatigue can also damage the next run, create longer transitions, and increase the chance of technical errors or no-reps.

This is why “running is king” cannot become “strength does not matter.” HYROX preparation needs enough strength to keep race loads manageable without adding unnecessary mass or fatigue that compromises running development.

What athletes and coaches should do now?

1. Test fresh running capacity

Begin with a reproducible fresh-running benchmark such as a 3-km or 5-km time trial. When resources permit, add VO₂max, velocity at VO₂max, LT1, LT2 or a defensible threshold proxy, and running-economy assessment. The goal is to establish whether the athlete is slow before fatigue is introduced.

2. Test HYROX durability

Use repeated 1-km efforts with standardized station inserts. Compare early and late pace, heart rate, RPE, mechanics, and recovery into the next run. A good fresh runner whose pace collapses after stations needs a different prescription from an athlete whose fresh running is already the primary limitation.

3. Establish the strength floor

Assess race-load sled push and pull, farmers carry, sandbag lunges, and wall-ball capacity with valid technique. The question is not only how strong the athlete is in the gym. It is whether race loads can be completed without excessive breaks, technical collapse, or a severe next-run penalty.

4. Measure efficiency, not only fitness

Track cadence, breaks, transition time, setup, no-rep risk, and the effort required to achieve each station split. The fastest possible SkiErg or row split is not automatically the fastest total-race decision if it causes a larger loss during the following kilometer.

5. Classify the primary limiter

• Fresh-run limited: running is weak before station fatigue.
• Durability limited: fresh running is good, but pace decays sharply after work.
• Strength-floor limited: race loads require excessive breaks or create extreme slowing.
• Skill/efficiency limited: transitions, technique, setup, or no-reps leak time.
• Pacing/fueling limited: aggressive early execution or late energy loss drives the fade.

The programming model: endurance-led, strength-protected

A sensible default is to make running the largest endurance exposure while preserving two strength touches and one controlled race-specific session. The exact volume depends on training age, injury history, division, race date, available time, and recovery capacity.

• Easy aerobic running builds volume, tissue tolerance, and economy at relatively low cost.
• Threshold-oriented training raises sustainable speed and metabolic control.
• Appropriate VO₂max work develops high-end aerobic power when the athlete profile justifies it.
• Heavy strength work protects force reserve and makes race loads a smaller percentage of maximum capacity.
• Compromised running teaches transfer, pacing, transitions, and mechanics under race-specific fatigue.

Important to consider: Do not turn every run into a compromised metcon. Fresh running develops qualities that fatigue can obscure; compromised running teaches transfer. Most athletes need both.

Retest the intended adaptation

After four to eight weeks, repeat the same benchmark under similar conditions. Success is not simply a faster simulation. Look for the mechanism you intended to change:

• Faster fresh running at similar or lower physiological cost.
• Smaller pace decay between early and late compromised kilometers.
• Fewer station breaks and less next-run disruption.
• Better execution without increased pain or excessive fatigue.
• Improved total-race modeling without losing the strength required for the division.

What the study cannot tell us

Race-result data do not include training history, VO₂max, lactate thresholds, VLamax, economy, body composition, injury status, footwear, environmental conditions, travel, sleep, or fueling. Participation and competitive density also changed dramatically across seasons, and Season 3 was disrupted by the COVID-era reduction in events.

The study therefore identifies where the broad performance signal lives. It does not prescribe an identical weekly plan for every athlete.

Coaching notes:

Running is the largest performance lever in HYROX — but it only remains available when the athlete can survive the stations efficiently. The best preparation is not running-only and it is not random high-intensity functional training. It is endurance-led, strength-protected, technically efficient, and individually tested.

The study tells us where to look. Your athlete data tells us what to train.

NEXT STEP Use Free Warrior Lab app and WPL free guide to begin identifying your performance limiter HYROX pathway to Test → Analyze → Train → Retest.

References:

Brandt, T., Ebel, C., Lebahn, C., & Schmidt, A. (2025). Acute physiological responses and performance determinants in HYROX© — A new running-focused high intensity functional fitness trend. Frontiers in Physiology, 16, Article 1519240. https://doi.org/10.3389/fphys.2025.1519240

Davids, C. J. (2026). A performance analysis of HYROX: A review of the physiologic, mechanical, and technical demands. Strength & Conditioning Journal, 48(1), 88–100. https://doi.org/10.1519/SSC.0000000000000913

Drum, S. N., Rappelt, L., & Donath, L. (2019). Trunk and upper body fatigue adversely affect running economy: A three-armed randomized controlled crossover pilot trial. Sports, 7(8), Article 195. https://doi.org/10.3390/sports7080195

Jones, A. M. (2024). The fourth dimension: Physiological resilience as an independent determinant of endurance exercise performance. The Journal of Physiology, 602, 4113–4128. https://doi.org/10.1113/JP284205

Rappelt, L., Wiedenmann, T., Held, S., Heinke, L., Micke, F., Wicker, P., & Donath, L. (2026). Longitudinal performance development in PRO and ELITE HYROX competitions across the first seven competitive seasons. Frontiers in Physiology, 17, Article 1847569. https://doi.org/10.3389/fphys.2026.1847569