Writing

HYROX comes to Navy Pier in February. We can train half of it.

HYROX is back in Chicago. The race is scheduled for 11–15 February 2027 at Navy Pier — the same hall that has hosted it before, on the lakefront, in February, which tells you something about the conditions in the roxzone.

That is far enough out to build for properly. Sixteen-ish weeks of training is the difference between finishing the thing and racing it.

Dates are as published on the HYROX schedule at the time of writing. Confirm on the official race page before you book flights — the schedule moves, and registration windows close fast.

The race, in full

Eight runs of 1 km, each one followed by a station. You run first, then work, then run again, eight times through. 8 km of running total, and none of it fresh.

Station Work Open men Open women
1. SkiErg 1,000 m
2. Sled push 50 m 152 kg 102 kg
3. Sled pull 50 m 103 kg 78 kg
4. Burpee broad jumps 80 m
5. Row 1,000 m
6. Farmers carry 200 m 2 × 24 kg 2 × 16 kg
7. Sandbag lunges 100 m 20 kg 10 kg
8. Wall balls 100 reps 6 kg to 10 ft 4 kg to 9 ft

Loads shown are the Open division. Pro is heavier, doubles and relay split the work. Check the current rulebook before you train to a number — the loads have been revised between seasons before, and they will be again.

Most Open finishers land somewhere between about 1:20 and 1:50. The clock does not stop when you leave the running lane, which is why the transition area has its own name and its own reputation.

The interesting part: it is the same race everywhere

This is what makes HYROX worth writing about on a site like this one.

Almost nothing in fitness is comparable. Your gym's "murph time" means nothing against someone else's. A marathon PR is a course, a day, a wind direction. But HYROX runs the identical format — same distances, same loads, same order — in every city it visits, all season, worldwide.

Which means your finish time is not a vibe. It is a measurement, with hundreds of thousands of other results to sit against. You can compare yourself to your own last race, to your age group, to the person who beat you in Dallas. Almost nothing else in this industry gives you that.

That is the same argument this app makes about your bodyweight: a number is only useful when you know what it can be compared against, and what it cannot.

So it would be a poor advertisement for that idea if we were vague about what the app does for you here. Here is the honest split.

What XYFitAI genuinely does for a HYROX build

The strength base, programmed properly. Three splits — push/pull/legs, upper/lower, or full body — at three, four or five days a week, built from barbell, dumbbell, machine and bodyweight work. It runs as a five-week block: four weeks where volume climbs from a starting point to roughly one and a half times it while your target reps-in-reserve drops from 4 to 1, then a fifth week that cuts volume to around 60% and backs the effort off. That deload is not optional and not a suggestion — it is built into the block.

Sled pushes and wall balls do not build themselves. Legs that can push a loaded sled at 50 m and still lunge 100 m under a sandbag get built with squats, split squats, step-ups and hinges, in the eighteen weeks before anybody touches a sled.

Adaptation that reads your actual recovery. The app learns your volume tolerance and recovery speed from what you log and what you report, scores your recovery each week from energy, soreness and joint pain, and pulls volume or calls a deload when that score drops. For a race build — where you are stacking running volume on top of lifting, which is exactly how people get hurt in January — a system that reduces lifting volume when you are run down is the single most useful thing in here.

Arriving at race weight without gutting your training. This is the strongest thing the app does and it matters more for HYROX than for most events, because you carry your bodyweight for 8 km and then push a sled that does not care what you weigh.

It does not guess your maintenance calories from a formula. It measures them: mean logged intake, minus the energy implied by how your weight trend actually moved, across the same window. The weight trend is a smoothed 7-day mean with a least-squares slope fitted through it, and it throws out an obviously mistyped weigh-in rather than letting it drag the line. Until there is enough data — it wants roughly ten days of intake, a two-week span and six weigh-ins — it says so, and leans on the formula in the meantime.

Practically: it tells you what you are actually burning across a heavy training block, not what a height-and-weight equation thinks you burn. Cut off the real number and you keep your sessions. Cut off a formula's guess and you find out in week eleven.

There is a higher-carb macro preference for exactly this reason, and a weekly calorie budget you can shape unevenly across the week if you want more food around your hard days.

What it does not do, and you should know before you download it

This is not a HYROX app. It has never heard of HYROX. Specifically:

  • None of the eight station movements are in the exercise library. No SkiErg, no rowing erg, no sled push or pull, no farmers carry, no sandbag lunge, no wall balls, no burpee broad jumps. The library is resistance training — barbell, dumbbell, machine, bodyweight. Any "row" you find in it is a back exercise, not the erg.
  • The structured plans contain no running. If you build a plan from a split template, you will get lifting and nothing else. Runs only appear if you use the AI plan builder and ask it for endurance work in your own words.
  • Cardio you log is recorded, not coached. You can enter distance, time and pace on an AI-generated plan, and it will save them. But the adaptation engine, the plateau detection and the volume analysis all read your lifting only. Nothing tells you your 1 km pace is improving. Do not expect the AI to progress your running, because it does not look at it.
  • No pacing tools, no interval timer. The only timer is a fixed 90-second rest clock. There is no work/rest interval function, no target-pace prescription, no splits, no race projection.
  • No watch, no heart rate, no GPS. Everything is entered by hand. There is no Apple Health, Health Connect, Garmin or Whoop integration, and no heart rate anywhere in the app.
  • Plateau detection watches the barbell. It tracks bench, squat, deadlift and overhead press. It cannot see a stalled SkiErg time.

So: your runs, your compromised-running work, your station practice and your roxzone rehearsal happen somewhere else. A track, a gym with the kit, a HYROX-affiliated box — Chicago has several.

How to actually use it alongside the race work

If you are building for February, a workable shape looks like this.

Put the lifting here — three or four days a week on a split, and let the five-week block do what it does, including the deload. Let the app pull your volume when your recovery score says to; that is the week your running will thank you for it.

Do your runs and your station work at the gym, on your own plan. If you want them recorded in one place, build your plan through the AI builder and describe the hybrid goal in your own words — that is the only path that will attach cardio fields to your sessions. Just hold the expectation correctly: you are logging that work, not being coached through it.

And weigh yourself daily from now until race week, so that by December the app has a real measurement of your maintenance rather than an estimate. That is the number that decides whether you can drop a few kilos before February without your last four hard weeks falling apart.

Sixteen weeks is enough. Start the lifting block now, and let the race be a measurement rather than a surprise.