If you train for Hyrox four times a week, your training load lives in two places that refuse to talk to each other. Your intervals sit in a run tracker that thinks a sled push is a walk break. Your station work sits in a lifting app that has no idea you just ran 1km at threshold before touching the bag. Neither one can tell you the thing you actually need to know: is my total Hyrox load going up, flat, or quietly cooking me?
That gap is a method problem before it's a tool problem. You can't fix load tracking by installing another app if you don't yet have a unit of measurement that covers both a 4 × 1km interval set and eight minutes of farmer's carries. This guide gives you that unit, then shows you how to log compromised running and station work so the two become comparable.
- Session RPE (RPE × minutes) is the only unit that prices a sled push and a 1km run on the same scale. Adopt it before you change tools.
- Adding the totals from two apps double-counts recovery and underestimates the fatigue of compromised work.
- A simulation is logged as ONE session with splits. Sixteen separate entries break the maths.
- Keep your acute:chronic ratio between 0.8 and 1.3. Flat load with rising RPE is under-recovery, not progress.
Why running apps and lifting apps both fail Hyrox athletes
The reason is structural, not cosmetic. A run tracker models continuous aerobic work: distance, pace, time in zone. A lifting app models discrete work: sets × reps × load. Hyrox is neither. It's compromised work, running under fatigue and stations performed at a running heart rate. Neither model captures transitions or the run/station interaction, which is precisely what makes the race hard.
The natural reflex is to add the two totals. It doesn't work, for two opposite reasons. First, you double-count recovery: your lifting app counts the 90 seconds between sets as session time, your run tracker counts the recovery jog as distance, and the sum inflates a volume that never existed. Second, you underestimate real fatigue: a 1km run taken straight off a sled pull costs far more than the same kilometre run fresh, and neither app knows it.
What actually counts as "training load" for a hybrid athlete
Direct answer: use session RPE (sRPE). Load = RPE out of 10 × duration in minutes. It's the method validated by Foster and colleagues in 2001, and it's the only one of the common options that gives a sled push and a kilometre of running a comparable price.
The alternatives each have a blind spot. Tonnage (sets × reps × load) can't price a run. Distance or time-in-zone can't price wall balls. TRIMP and HRV-derived metrics are over-engineered for most age-group athletes: they need a chest strap worn consistently and a model few people maintain for three months. sRPE needs one number per session and survives a full year.
One detail that changes everything: take the RPE about 30 minutes after the session ends, not during. Taken in the moment, the number reflects the last effort rather than the whole session, and you systematically over-rate any session that finishes on wall balls.
Worked example: one Hyrox week scored end to end
| Session | Duration | RPE | Load |
|---|---|---|---|
| Intervals 5 × 1km | 55 min | 8 | 440 |
| Strength stations | 70 min | 7 | 490 |
| Half simulation | 45 min | 9 | 405 |
| Easy run | 50 min | 4 | 200 |
| Weekly total | 220 min | - | 1 535 |
The following week at 1,690 units is +10%, a healthy build. At 2,150 units it's +40% in seven days, and that's the profile Gabbett links to a marked rise in injury risk. The absolute number tells you almost nothing; the week-on-week step carries the signal.
How to log the three Hyrox session types so they stay comparable
Three conventions are enough, and they fix 90% of tracking inconsistency.
Interval runs: log total session time including recoveries, not just working time. A 5 × 1km with 3-minute recoveries is a 55-minute session, not 20 minutes of effort. Counting working time only underestimates load by roughly 60% and makes your interval weeks look artificially light.
Station work: record the station, the implement load, the reps or distance, and the time under load. A sled push is priced as distance × load, wall balls as reps × target height. Without the implement load you'll never explain why the same session cost RPE 6 in March and RPE 8 in May.
Race simulations: one session with splits, never eight runs and eight stations logged separately. This is the most violated and most costly rule. Sixteen entries erase transitions, which often run 4 to 8% of total time, and delete the single most useful signal in the session: which station degrades the run that follows it.
If you're planning a simulation, our Hyrox simulation workout guide walks through the full protocol and the splits worth recording.
The four ways athletes do this, scored honestly
| Method | Single timeline | Transitions | Auto weekly load | Admin / week |
|---|---|---|---|---|
| Two apps + mental maths | No | No | No | ~0 min |
| Spreadsheet | Yes | Yes | Manual | ~10 min |
| Generic all-in-one tracker | Yes | No | Partial | ~3 min |
| Hybrid-native tracking | Yes | Yes | Yes | ~1 min |
Be honest about the trade-off: the spreadsheet wins on flexibility and loses on adherence. It does everything you want, and most athletes abandon it within three to four weeks because ten minutes of Sunday-night data entry doesn't survive a hard block. Two apps plus mental maths is free and breaks even faster. The real selection criterion isn't feature depth, it's what you'll actually sustain for twelve straight weeks.
This guide covers method, not tool choice. If "which app" is what brought you here, our Hyrox app comparison scores them across seven criteria, and the hybrid buyer's guide covers the wider case.
How to read your load trend before a race block
Once you have a reliable weekly number, three readings are enough to steer a block.
The acute:chronic ratio. Divide this week's load by the average of the last four. Between 0.8 and 1.3, you're in a sustainable progression zone. Above 1.5, you're on the sharp-step profile associated with injury risk. Below 0.8 for several weeks running, you're losing fitness without noticing.
The taper rule. In race week, cut load by roughly 40% but keep the intensity. In practice: fewer and shorter sessions, not softer ones. Dropping to RPE 5 across race week gets you to the start line fresh and flat.
Flat load, rising RPE. This is the most important and most ignored signal. If you're doing the same work as three weeks ago and it costs you two more RPE points, you aren't progressing, you're under-recovered. The answer is a deload week, not another session.
Those three readings reduce to three decisions: add a session, cut a session, or hold. That's all a load number needs to produce. If it never changes a decision, stop calculating it.
Closing
Tracking Hyrox load needs neither a chest strap nor a complicated model. It needs a common unit, three logging conventions held honestly, and a weekly reading that ends in a decision. Session RPE covers all three. Start this week: record an RPE after every session, multiply by duration, and in a month you'll have the rolling average that finally makes "am I doing too much" an answerable question.
To place these numbers inside a full block, the 12-week Hyrox training plan gives the structure, and combining strength and running explains the physiology behind compromised work. For your interval paces, run the Hyrox pace calculator.
- Foster C. et al. (2001). A new approach to monitoring exercise training. Journal of Strength & Conditioning Research, 15(1), 109-115.
- Gabbett T. J. (2016). The training-injury prevention paradox: should athletes be training smarter and harder? British Journal of Sports Medicine, 50(5), 273-280.
- Impellizzeri F. M., Marcora S. M., Coutts A. J. (2019). Internal and external training load: 15 years on. International Journal of Sports Physiology and Performance, 14(2), 270-273.
- Hyrox official race format and station specifications, hyrox.com.
