One specific training session improved VO2max by 8.7%. The other group, doing the exact same training programme, only improved by 4.6%. Same sessions, same effort, nearly double the results. The only difference: one number on the bike computer screen. Cadence.
The coaches behind riders like Pogacar, Froome, and Bernal have been prescribing this exact session for years. While the internet and Reddit forums kept saying it was a waste of time, that it would hurt your knees. A brand new study has just proved those coaches right.
Why This Study Matters
This argument has been raging in cycling forums for decades. On one side, you have the old school coaches and the professional peloton — people who've been prescribing big gear work since before power meters existed. They'd send riders to find a long climb, slap it into the 53-11, and grind at 50 RPM for 10-minute intervals.
On the other side, the science-only crowd. And the research was on their side — until now. Previous studies (Christopherson 2014, Nimmer 2012, Luda 2016, Witty 2016) found no difference or even favoured freely chosen cadence. The internet had its ammunition.
But there was a major problem with all those studies. The protocols didn't match what the coaches were actually doing. The cadences were wrong. The intensities were wrong. The durations were wrong. It was like testing weightlifting by getting people to lift 2kg dumbbells, seeing no reaction, and concluding that strength training doesn't work.
The coaches knew the protocols were wrong. So they kept prescribing these sessions anyway and waited for the science to catch up.
The 2024 Habis Study
In November 2024, researchers Raphael Habis and Paulina Habis from the Rocklaw University in Poland published a study in PLOS ONE. They took 24 well-trained female cyclists — experienced riders with at least three years of competitive cycling, training 10+ hours per week — and split them into two groups on identical 8-week polarised training plans.
The only difference: one group did their intervals at freely chosen cadence (~80 RPM). The other did them at 50-70 RPM.
The Results
VO2max — the single most important marker of aerobic fitness:
- Low cadence group: +8.7%
- Freely chosen cadence: +4.6%
- Statistically significant (p = 0.02)
Maximum aerobic power — the absolute ceiling of your aerobic engine:
- Low cadence group: +8.1%
- Freely chosen cadence: +3%
- Statistically significant (p = 0.03)
And one detail that jumped out: the freely chosen cadence group gained an average of 0.7kg of body mass. The low cadence group's body mass stayed exactly the same.
Why It Works: Three Mechanisms
1. Greater Muscle Fibre Recruitment
At high cadence, your slow-twitch (Type 1) fibres handle most of the work. Drop the cadence and increase the torque demand, and your body recruits bigger, more powerful Type 2 fast-twitch fibres. Over a sustained training block, those fast-twitch fibres begin to shift — developing more mitochondria, better blood supply, improved oxidative capacity, while retaining their force-generating ability. You're expanding your aerobic engine by putting previously underused muscle fibres to work.
2. Neuromuscular Pathway Development
When John Wakefield from Red Bull–Bora–Hansgrohe was on the podcast, this is exactly what he described. Pedalling at low cadence under high force trains the nervous system to activate more muscle fibres simultaneously and coordinate that activation more efficiently.
In a race, four hours deep, your cadence is going to drop from fatigue. Riders who've trained their neuromuscular pathways through torque work can produce meaningful power at that reduced cadence. Riders who haven't just fade.
3. Improved Gross Efficiency
Gross efficiency is how much power you get out of each calorie you burn. By training under high torque conditions, you teach your muscles to produce force more economically. When you go back to riding at normal cadence, you carry that improved efficiency with you. Same effort, more power. Same power, less energy cost.
The Exact Sessions to Do This Week
These come from John Wakefield — director of coaching at Red Bull Red Bull–Bora–Hansgrohe, one of the most respected performance minds in the WorldTour.
Session 1: 4-Minute Torque Intervals
- Find a climb, 4-7% gradient
- 4 minutes at 40-60 RPM
- RPE 7/10 — controlled, deliberate, not all-out
- 4 minutes easy spinning recovery
- Start with 3 reps, build to 5 over several weeks
Session 2: 10-Minute Torque Intervals
- Same gradient
- 10 minutes at 40-60 RPM
- Equal rest between intervals
- Start with 2 reps, build to 4
The 8-Week Progression
Weeks 1-2: Start at 65 RPM. 3 sets of 4 minutes. Moderate effort. Getting joints and connective tissue accustomed to increased force.
Weeks 3-4: Drop to 55-60 RPM. 4 sets. Effort up to 7/10.
Weeks 5-6: Push toward 50 RPM if joints feel good. Introduce 10-minute format.
Weeks 7-8: Full protocol. Do this twice per week during base/build phase.
Critical Warning
Torque training puts enormous stress on knees, tendons, and connective tissue. If you jump into 40 RPM efforts without building up gradually, you will get injured. Inside the Not Done Yet coaching community, we see riders who've executed this badly and we're picking up the pieces. Start conservative. Complement with gym-based strength work.
Do these sessions outdoors on a real climb if possible. If you're new to climbing at intensity, start with shorter gradients. The gradient provides natural, consistent resistance. On a trainer in ERG mode, riders often report the feel is fundamentally different — the ERG gets on top of you at low cadence.
The Bigger Lesson
For years, the cycling internet has been obsessed with the idea that science trumps coaching experience. In many cases, that's right — we should demand evidence. But what this study shows is something the best coaches have always understood: sometimes the science takes a while to catch up to what works in practice.
Key Takeaways
- Low cadence interval training (50-70 RPM) produced 8.7% VO2max improvement vs 4.6% at freely chosen cadence
- Previous studies that found no benefit used protocols nothing like what coaches actually prescribe
- Three mechanisms: greater muscle fibre recruitment, neuromuscular development, improved efficiency
- Start at 65 RPM and build down gradually over 8 weeks to protect your joints
- Do these twice per week during base/build phase, ideally on a real climb
- Always complement with gym-based strength work
- This is not a replacement for Zone 2 or VO2max intervals — it's an additional tool
- Always complement with gym-based strength work
- For the full cadence discussion, see our optimal cadence guide
- Use the FTP Zone Calculator to find your exact power targets for torque intervals
- Protect your knees by progressing gradually over 8 weeks
