Heavy or high-rep lifting to help at a station?
27 cadets, 22 weeks, 10RM vs 30RM to failure: 10RM gained more strength and jump; no clear difference in leg-extension reps at a fixed load. No station tested; not hybrid athletes.
Main caveat: One endurance test, one joint. Leg extension at a cadence is far from a loaded station under lactate; its result has no number in the text.
Øfsteng and colleagues compared high load (HL, 10RM) and low load (LL, 30RM) whole-body resistance training in 27 cadets of the Norwegian Defence Cyber Academy. Participants were pair-matched on initial strength and then randomly allocated (HL 14, LL 13).
Both groups did three sets of seven exercises to concentric failure, twice a week for 10 weeks, then after a three-week leave three days a week every other week for 9 weeks, 37 sessions in all. HL trained at 75.5% of 1RM and LL at 50.8%, and LL lifted 25119 kg a week against 13687 kg.
At week 22, HL had gained more in the combined upper-limb and lower-limb strength scores, jump height and arm lean mass. Agility, muscle endurance, leg lean mass and vastus lateralis fibre area showed no difference. The muscle endurance test was leg extension to failure at 60% of pre-test body mass, at a metronome cadence, with the same absolute load at every time point.
Our reading: a clean, long and supervised answer for strength and jump in moderately trained young adults, and one modest data point that training with 30RM did not out-gain 10RM on one fixed-load endurance test.
It does not answer "heavy or high reps for stations": the test is single-joint, its size is only in a figure, and 10RM to failure is not the heavy, short-of-failure lifting the base recommends.
Protocol
Design: randomised, repeated-measures, parallel; n = 27 (HL: male 12, female 2; LL: male 10, female 3). Testing at weeks 0, 10 and 22 after a familiarisation session. Test personnel were blinded to allocation. No non-training control.
Who: 20 ± 1 years, 75.5 ± 12.9 kg, 182 ± 9 cm; 2nd-year cadets. Excluded: more than 2 systematic heavy sessions a week in the prior six months.
Training: seven exercises in fixed order (squat, leg press, leg curl, bench press, standing rowing, seated pull-down, seated biceps curl), three sets, 2-3 min between sets, every set to concentric failure, load raised to stay at the target RM. 20 sessions in the first 10 weeks, a three-week leave, 17 sessions in the last 9 weeks, supervised. 30 g whey after each session.
Other load: 1-2 weekly mandatory 60-min academy sessions and a two-week field exercise; additional endurance training 9.3 ± 7.7 h (HL) and 7.5 ± 7.3 h (LL) over the study.
Muscle endurance test: leg extension at 60% of pre-test body mass, repetitions to failure, 2 s concentric and 2 s eccentric to a metronome, stopped when the cadence was missed two consecutive times; the same absolute load at all test time points.
Other outcomes: 1RM biceps curl, bench press and leg press; maximal isometric half squat; counter movement jump; agility run; DXA lean mass; vastus lateralis biopsy; fasting hormones.
Analysis: linear mixed models on relative change, baseline and sex as covariates; no correction for multiple outcomes is described; alpha 0.05. No sample-size calculation is reported.
Funding and interests: supported by the Inland Norway University of Applied Sciences and the Norwegian Defence Cyber Academy (which also supplied the participants and some authors); no conflicts of interest declared. Pre-registered in the Norwegian Center for Research database (project number 43901/3), as printed.
Key Results
Between-group mean difference at week 22 (HL minus LL, percentage points, 95% CI):
| Outcome | Difference | CI | P |
|---|---|---|---|
| Upper-limb strength, combined | 10 | 2.8 to 17.1 | 0.010 |
| Bench press 1RM | 7.8 | 1.6 to 14.1 | 0.016 |
| Biceps curl 1RM | 9.1 | 0.1 to 18.1 | 0.049 |
| Lower-limb strength, combined | 9.9 | 1.1 to 18.6 | 0.029 |
| Leg press 1RM | 14.7 | 3.9 to 25.6 | 0.016 |
| Maximal isometric half squat | 7.1 | -2.3 to 16.5 | 0.130 |
| Counter movement jump | 5.5 | 1.4 to 9.6 | 0.011 |
| Arm lean mass | 5.2 | 1 to 9.4 | 0.018 |
| Leg lean mass | 1.3 | -1.6 to 4.1 | 0.374 |
| Pooled fibre cross-sectional area | 4.0 | -14.7 to 22.7 | 0.656 |
| Agility | no difference | Figure 4B only | not printed |
| Muscle endurance (repetitions) | no significant difference; about +12 favouring HL, by eye from Figure 4C (our reading) | about -8 to +32, by eye | not printed |
Training actually done: HL 75.5 %1RM [71.7, 79.2], LL 50.8 %1RM [47.2, 54.5]; weekly volume load 13687 kg [12324, 15049] vs 25119 kg [23720, 26518]; adherence 92% (SD 8) vs 96% (SD 7).
Over time, both groups: counter movement jump, agility and muscle endurance improved (time effects P<0.001, P=0.0268, P=0.004).
Timing: the upper-limb and jump advantages appeared by week 10 (e.g. upper-limb strength 10.3, CI 4.4 to 16.1), with none added afterwards.
Mechanisms
Measured: HL gained more strength than LL in both limbs and more lean mass in the arms; leg lean mass and fibre area did not differ, so leg strength and leg mass decoupled between loads. Hormones were similar throughout; type IIX fibres fell in both groups.
Authors' interpretation: part of the 1RM advantage may be skill from training closer to the test (the isometric half squat did not differ). The upper-lower difference may reflect prior training status of the legs, concurrent aerobic work, or muscle biology; they call these explanations speculative.
Inference: the endurance test used a fixed external load, so a stronger person meets it at a lower share of their maximum, as REF-1509 discusses. HL got stronger and LL trained with far more reps; the two routes ended with no significant difference on this one test, with a wide interval.
Practical Application
Heavy or high-rep lifting to help at a station? 27 cadets, 22 weeks, 10RM vs 30RM to failure: 10RM gained more strength and jump; no clear difference in leg-extension reps at a fixed load. No station tested; not hybrid athletes.
What it is good for: as a long, supervised comparison in moderately trained young adults. For strength and jump, the heavier arm won on about half the weekly volume load. On one fixed-load endurance test, the extra volume of 30RM showed no advantage, with a wide interval.
What it does not show: which load improves time at a sled, a lunge or wall balls; anything in athletes who already lift or run; or anything about lifting short of failure. Every set here was to concentric failure, and 10RM at 75.5 %1RM is not SYNTH-44's heavy work.
Not from this paper; the base's own practice: keep heavy strength short of failure as the default in a HYROX build, with one station session a week turning into race-specific combinations in the last 6-8 weeks (SYNTH-71); that guide lists high-repetition vs heavy work for station time as a gap, not a rule.
Limitations
One endurance test, one joint. Leg extension at a cadence is far from a loaded station under lactate; its result has no number in the text. Figure 4C, read by eye (ours, not printed values), shows about 37 to 49 repetitions in HL and 39 to 46 in LL, and a week-22 difference of about +12 percentage points favouring HL with an interval of about -8 to +32: compatible with a sizeable gain for HL or a small one for LL.
Not powered for a null. No sample-size calculation is reported; 27 participants split in two cannot show that the groups are equal on endurance or leg mass.
Possible attrition, only in a footnote. Table 1's footnote on the week-22 body-composition rows gives high-load male n = 11, low-load female n = 2, male n = 9 (24 of 27, our sum, if the 2 high-load women stayed); the text gives no flow of participants and reports no dropouts.
Not volume-matched, no control group, as the authors note. Military stressors (field exercise, leave, energy deficit) were shared but may interact with the training.
Multiple outcomes with no correction; biceps curl at P=0.049 would not survive one. The reported effect size for the jump, 0.60 [-0.20, 1.38], crosses zero while its mean difference does not.
Small inconsistencies. The abstract says the differential responses appeared "primarily" in the first ten weeks for all variables, but the lower-limb advantage accumulated gradually with neither phase significant. Methods give more than 4 hours between same-day sessions; the discussion says at least 12 h.
Women: 5 of 27; no separate analysis shown.
Source used: accepted manuscript, not the typeset version.
How this entry was checked and changed
Why this level
parallel RCT with 27 participants; randomisation confirmed in the full text; Table 1's footnote implies 24 at week 22 (our sum); the endurance result is a single-joint test with no number in the text
Full text (checked 2026-10-30)
Participants were pair-matched based on initial strength performance and then randomly allocated to a heavy-load group (male: 12, female: 2) or a low-load group (male: 10, female: 3) (full text)
Muscle endurance: leg extension at a load corresponding to 60% of pre-test body mass, cadence 2 seconds in both concentric and eccentric phase; the same absolute load at all test time points (full text)
HL 75.5 %1RM vs LL 50.8 %1RM; training volume 13687 kg vs 25119 kg per week (full text)
For agility and muscle endurance, no differences were observed between training modalities (full text)
Time effects for jump, agility and muscle endurance: P<0.001, P=0.0268, P=0.004 (full text)
Lower-limb weighted strength score: 9.9 % mean difference, [1.1, 18.6], P=0.029; arm lean mass 5.2 % mean difference, [1, 9.4], P=0.018 (full text)
Supported by the Inland Norway University of Applied Sciences and the Norwegian Defence Cyber Academy; no conflicts of interests declared (full text)