Hyrox Training Load: How Running Volume and Station Work Stack Up

Hyrox is eight kilometres of running with eight strength stations wedged between them. That structure is the whole appeal, and it is also why prep for it goes wrong in a specific, predictable way.

To race it well you have to build two things at once: enough running volume to hold pace across eight separate kilometres, and enough strength and work capacity to get through sleds, carries, lunges and wall balls without the runs afterwards falling apart. Most prep blocks progress both. Both progressions look sensible on their own. The total does not.

Why a Hyrox block ramps twice

Take a fairly normal build. You add ten percent to your weekly running over three weeks, which is unremarkable. Over the same three weeks you add a second station session and start pushing the sled heavier, which is also unremarkable.

Neither decision is wrong. But you have two independent progressions landing on one body, and nothing you are looking at adds them together. Your running app shows a sensible running build. Your gym log shows a sensible strength build. The thing that actually accumulated is the sum, and no chart on your phone is drawing it.

This is the single most common way a Hyrox block gets away from someone: not one bad decision, but two reasonable ones that were never counted together.

What your watch sees, and what it misses

Heart-rate-based load has a specific blind spot, and Hyrox stations sit right in it.

Sled pushes, sled pulls, farmers carries and lunges are heavy, short and grip-limited. They tax the posterior chain and the anaerobic systems hard, but they do not hold your heart rate high for long enough to look expensive to a model built on heart rate over time. Wall balls and burpee broad jumps score better because they run longer, but still under-represent what they cost you.

So the session where you did sleds and carries gets logged as moderate, sometimes as recovery. Two days later your legs disagree with that assessment during an easy run, and nothing in your data explains why.

If you want the longer version of that problem and the simplest honest fix, the session-RPE method covers it.

The number that actually predicts your race

Here is the thing worth measuring, and almost nobody does it.

After a simulation, most people look at total time and at which stations were slow. Both are useful. But the number that tells you the most about your race is how much your running degrades from the first half to the second.

Pull only the runs out of your sim. Work out the per-kilometre pace of each one. Then compare the average of the first four against the average of the last four.

An illustrative case: someone whose opening run is at 5:00 per kilometre and whose final run is at 6:10 has slowed by 23 percent over the course of the session. Averaged into halves, their first four runs sit around 5:15 and their last four around 5:55, a fade of roughly 12 percent.

Why this matters more than the station splits: across a full race the runs are the majority of your moving time. A twelve percent decay applied across four kilometres costs far more than tightening up a transition or shaving fifteen seconds off the farmers carry. And because fatigue compounds rather than adding up neatly, doubling a half-simulation almost always reads optimistic.

What to do with that number

If your fade is large, the training target is not the sleds. It is the specific ability to run at a controlled pace while already deep in accumulated fatigue.

• Pace the opening runs slower than feels right. Going out at a pace you cannot hold means paying it back with interest across the back half. Flat splits beat fast-then-fading almost every time.
• Practise compromised running at the end of long sessions, not fresh. A hard interval session while rested rehearses a state that never occurs in the race.
• Rehearse transitions. They are free time and they are the cheapest thing on this list to improve.

Phase the two sides, do not stack them

The practical fix for the double ramp is boring and it works: stop progressing both at once.

Over a ten to twelve week block, alternate which side is allowed to grow. Spend three weeks adding running volume while holding station work steady at whatever you can already handle. Then hold running flat and spend three weeks progressing the stations — heavier sled, more volume on carries, longer sets of wall balls. Then repeat, and let the last couple of weeks come down.

You still arrive at the race having built both. What you avoid is the stretch where both are climbing simultaneously and your total is ramping at twice the rate either individual plan implies.

Reading it as one number

All of this gets much easier if you have a single figure covering both sides, because then the ramp is visible rather than inferred.

Once you have that, the useful thing to watch is not any single week's total but how fast it is changing relative to your recent baseline. That is what the acute-to-chronic ratio describes, and it is worth understanding both its logic and its real limitations — the honest version is here. Treat it as a ramp-rate signal that tells you when to pay attention, not as a prediction, and certainly not as medical advice.

For a Hyrox block specifically, the ratio is only meaningful if your station work is actually in the load figure it is built on. If it is not, you have a running ratio that will sit calmly at 1.0 through the exact weeks you most needed a prompt.

Disclosure

I built LOAD for this problem. It pulls your runs from Strava or Polar and your lifting from Hevy or manual entry, fuses them into one weekly number, and tracks the ramp rate over the combined figure rather than the cardio half. It is free for individual athletes, and connecting Strava backfills 12 weeks so the baseline is real before your block starts.

None of the analysis above needs it, though. The fade calculation is four numbers and a calculator, and phasing your progressions costs nothing at all.

The methodology and citations are public at loadtraining.app/faq.