Key Takeaways (30-second version)
- %1RM standardizes the weight, not the stimulus. The same percentage produces wildly different repetition counts in different people.
- The lower the percentage, the wider the gap. That is exactly where BFR works (20-30 %).
- Pressure has the same problem. A fixed number, whether %1RM or % AOP, is a starting point, not a prescription.
- Steer by response, not by table: individually calculated pressure, perceived tightness as feedback, documented sessions.
1. The problem hiding in plain sight
Open almost any resistance training study and you will find the same sentence: “Participants trained at (e.g.) 70 % of their one-repetition maximum (1RM).” It sounds precise. It looks like a standardized dose.
It isn’t. Only the external load, the mass in kilograms, is standardized. What that load does to the person lifting it differs from one individual to the next. In BFR this matters even more, because BFR protocols typically prescribe just 20-30 % of 1RM.
This article makes three points. First, %1RM is a weaker control variable than the field assumes. Second, the weakness grows as the load gets lighter, which is exactly where BFR operates. Third, the same pattern appears with cuff pressure, and there is a practical alternative.
2. What the data show
Buskies and Boeckh-Behrens (1999) asked a simple question: how many repetitions can people perform at a given percentage of their 1RM? The study was published in Leistungssport, a German journal, before the “publish or perish” era. Today, a paper from 1999 is easily dismissed as “old.” That is a mistake. Work like this was done slowly, checked critically and reported in detail, including the minimum and maximum values of individual participants. Most international readers have never seen it. Old but gold.
The finding is clear. People lifting the same percentage of their 1RM achieve very different repetition counts.
Table 1: Repetitions at the same %1RM (range across participants; Buskies & Boeckh-Behrens, 1999)
| Exercise | 90 % 1RM | 50 % 1RM |
|---|---|---|
| Leg press, 90° knee angle (n = 16) | 3-18 | 26-99 |
| Leg press, 45° knee angle (n = 16) | 3-16 | 21-100 |
| Triceps press, both arms (n = 16) | 2-19 | 35-89 |
| Bench press (n = 14) | 4-10 | 22-41 |
| Lat pulldown (n = 14) | 4-7 | 29-51 |
At 90 % of 1RM, one person managed two repetitions in the triceps press, another nineteen. At 50 %, the leg press ranged from about 21 to 100 repetitions. The authors give a vivid example: at 70 % of 1RM in the leg press, one sport student completed 13 repetitions, another 61. The first is closer to muscle-building training, the second closer to endurance training. Same percentage, completely different stimulus.
Same percentage, up to ten times the repetitions
At the same percentage of their one-repetition maximum, participants in Buskies and Boeckh-Behrens (1999) achieved repetition counts that differed by hundreds of percent:
- Triceps press at 90 % 1RM: 2 to 19 repetitions, a difference of +850 %
- Leg press at 90 % 1RM: 3 to 18 repetitions, a difference of +500 %
- Leg press at 70 % 1RM: 13 vs. 61 repetitions in two individual students, a difference of +369 %
- Leg press at 50 % 1RM: 21 to 100 repetitions, a difference of +376 % (nearly 80 repetitions apart)
Percent difference = (highest − lowest) ÷ lowest × 100. Data: Buskies & Boeckh-Behrens (1999), Tables 3 and 4.
The takeaway: “Same %1RM” does not mean “same stimulus.”
3. The scissors open downwards
Look at Table 1 again. At 90 % the spread is large. At 50 % it is far larger. The lower the percentage, the wider the gap, and that is the range BFR works in.
Two more sources point the same way at low loads:
- Jenkins et al. (2015) had participants perform leg extensions to failure at 30 % and 80 % of 1RM. The study was not a BFR study; it shows how repetitions behave at light loads, nothing more.
- Nuzzo et al. (2024) pooled many studies in a meta-regression. The number of repetitions at a given %1RM depends on sex, age, training status and exercise.
The step to BFR is therefore an inference, but a sensible one: if variability already grows without cuffs, a protocol that prescribes “30 % 1RM” cannot guarantee a comparable stimulus between participants.
4. Why BFR makes the problem worse
At 20-30 % of 1RM the load itself contributes little. Other variables take over: cuff pressure, number of sets and repetitions, rest intervals, tempo and proximity to failure. Yet studies still report the load as if it were the controlling variable.
The consequence: two participants in the same group, with the same “30 % 1RM,” can experience very different effects. Research averages these differences and concludes that “BFR at 30 % works” or “doesn’t work.” The average hides the individual.
5. The same mistake with pressure
BFR research moved from absolute cuff pressures to a percentage of arterial occlusion pressure (AOP) to make results comparable. That was real progress. But the widely cited 40-80 % AOP range turned into a “recommendation” step by step: studies used it, reviews described it, later papers cited it. No single study had established 40-80 % as optimal (read the full analysis).
This is our interpretation of how the literature developed, not a proven fact. But the pattern is the same as with %1RM: a number that is easy to report is mistaken for a number that controls the stimulus.
6. What Toigo and Boutellier actually tell us
Toigo and Boutellier (2006) did not run an experiment. In their review they described the determinants of resistance training that drive molecular and cellular adaptation. The message: load times repetitions is only a fraction of the picture.
Table 2: Same “load,” different stimulus
| Determinant | Why it matters in practice |
|---|---|
| Time under tension | Same repetitions, very different workload |
| Tempo | Changes mechanical and metabolic demand |
| Rest between repetitions | Determines how much fatigue builds |
| Proximity to failure | Strongest driver of actual effort |
| Range of motion | Alters muscle length and work |
| Recovery between sets | Determines accumulated fatigue |
For researchers, the point is to report all of these, not just kilograms. For practitioners it is simpler: if you don’t know these variables, you don’t know your stimulus.
7. Testing the untestable
There is also a practical problem. Maximal strength tests are a questionable tool for older adults, people with joint problems or patients in rehabilitation. Buskies and Boeckh-Behrens themselves note that the test is problematic. This point rests on practical experience, not on a literature review. If the people who need training most cannot be tested reliably, why prescribe their training by a test they cannot do?
8. What to do instead: structured individual autoregulation
The solution is not a better table. It is a loop: input, output, feedback, correction.
Step 1: Start from the person, not the menu. Dr. Yoshiaki Sato developed “optimal pressure” by experiment, after a pulmonary embolism in 1967 showed him what happens when pressure is pushed too far. His question was never “how much can I tolerate” but “what is the appropriate pressure” (Why Sato Invented Optimal Pressure).
Step 2: Calculate a starting hypothesis. The KAATSU Pressure Calculator uses limb circumference, age, blood pressure, resting heart rate and KAATSU experience. For the example in that article (arm, 52-year-old man, 31 cm, 123 mmHg, 55 bpm) it returns an upper limit of 160 SKU. That is a starting point, not the answer.
Step 3: Start low, progress in small steps. Begin at about 80 SKU for arms (200 SKU for legs), increase by 10 SKU per interval and stop at the recommended ceiling.
Step 4: Feed the response back. The user rates perceived band tightness from 0 to 10. In the example, a 4 out of 10 (“noticeably below target”) moved the recommendation to 170 SKU. The system treats the user’s response as data.
Step 5: Choose the mode.
- Constant Mode: pressure stays on, including during rest. The practitioner decides inflation duration, set structure and rest.
- Cycle Mode: the device alternates inflation and deflation in preset intervals, for example 30 seconds on and 5 seconds off over 8 steps. Time replaces rep counting.
Step 6: Stop at the efficacy threshold. The target is the point where metabolic stress is first felt, not the highest pressure tolerated. As much pressure as necessary, not as much as possible.
The Calculator does not replace manual tightness checks, a trained eye or professional supervision. What it does is keep the user inside the loop of judgment instead of removing them from it.
Table 3: From %1RM to autoregulation
| Problem | Autoregulated alternative |
|---|---|
| Same % gives different reps | Effort or fatigue target, e.g. declining reps across sets |
| 1RM test unsafe or unreliable | No maximal test needed |
| Fixed % AOP or fixed pressure | Individually calculated pressure, adjusted by feedback |
| Stimulus undocumented | Record tempo, rest, steps, SKU and perceived tightness |
9. Science is not the enemy
This is not a rejection of research. Toigo and Boutellier themselves ask for more precise reporting. The issue is that group studies deliver averages, while training happens to individuals. Research cannot, by its structure, deliver the fine-tuning for the person in front of you. That is not common sense replacing science. It is structured, documented individualization, informed by what research does show.
Key Takeaways (checklist)
Before your next session, ask yourself:
- Do I know what stimulus this load actually produces for this person?
- Is my pressure calculated and adjusted, or copied from a table?
- Did I document tempo, rest, repetitions and perceived tightness?
- Can this person be tested safely? If not, why prescribe by a test they cannot do?
Bottom line: Weight and pressure are inputs. The response is the dose.
References
- Buskies, W., & Boeckh-Behrens, W. (1999). Probleme bei der Steuerung der Trainingsintensität im Krafttraining auf der Basis von Maximalkrafttests. Leistungssport, 29(3), 4-8.
- Jenkins, N. D. M., Housh, T. J., Bergstrom, H. C., Cochrane, K. C., Hill, E. C., Smith, C. M., Johnson, G. O., Schmidt, R. J., & Cramer, J. T. (2015). Muscle activation during three sets to failure at 80 vs. 30% 1RM resistance exercise. European Journal of Applied Physiology, 115(11), 2335-2347. https://doi.org/10.1007/s00421-015-3214-9
- Nuzzo, J. L., Pinto, M. D., Nosaka, K., & Steele, J. (2024). Maximal number of repetitions at percentages of the one repetition maximum: A meta-regression and moderator analysis of sex, age, training status, and exercise. Sports Medicine, 54(2), 303-321. https://doi.org/10.1007/s40279-023-01937-7
- Sato, Y. (2005). The history and future of KAATSU Training. International Journal of KAATSU Training Research, 1(1), 1-5.
- Toigo, M., & Boutellier, U. (2006). New fundamental resistance exercise determinants of molecular and cellular muscle adaptations. European Journal of Applied Physiology, 97(6), 643-663. https://doi.org/10.1007/s00421-006-0238-1