Low cadence / torque training (“K3”, “Big Gear”) 🚴♂️⚙️
What the studies of the last 20 years really show – and how you as a cyclist or triathlete can meaningfully incorporate it.
You know them: these units with a high gear and low cadence – 50–70 rpm, sometimes even 40 rpm. In German-speaking countries often called “K3”, internationally “torque training” or “low-cadence training”.
The idea sounds logical: more torque = more "power endurance" at the pedal .
But: Logical ≠ effective . And the research was, for a long time… let's say: contradictory. 😅
The current PLOS ONE study (2024) has given the topic a real boost – especially because it tests torque training in a polarized setup and shows clear effects in well-trained female athletes.
1) What exactly is "torque training" from a physiological perspective? 🧠🦵
For the same performance:
Power = Torque × Angular velocity (cadence)
This means that if you lower your cadence (e.g., from 90 to 60 rpm), the torque has to increase to produce the same wattage. This shifts the workload more towards the muscular/neuromuscular system (more power per pedal stroke) – often with a similar cardiovascular load, depending on the setup.
Figure 1 shows exactly that (at 250 W): low cadence = significantly higher torque.
2) 20 years of research: What has changed? 📚🕰️
Phase 1: “Optimal Cadence” & Efficiency (2000s)
Early work (e.g. Foss & Hallén; Lucia et al.) has shown:
The freely chosen cadence is often higher than the cadence that would be most mechanically/energetically efficient (classical paradox).
This led to much speculation: “If a lower cadence is more efficient, shouldn’t we train that…?” These classics are older than 10 years, but remain relevant as a foundation (and are still cited today, e.g. in Kristoffersen 2014).
Phase 2: First intervention studies (2009–2016) – mixed results
Paton et al. (2009) : Low-cadence intervals (60–70 rpm) sometimes showed better performance/physio effects than high-cadence intervals – but the study is older and the protocols were not always “cleanly” controlled.
Nimmerichter et al. (2012) : Low-cadence intervals (60 rpm) – especially uphill – improved time trial power (significantly uphill, slightly on flat terrain). DOI: 10.1007/s00421-011-1957-5 .
Kristoffersen et al. (2014) (veterans, 47±6 years): 12 weeks at 40 rpm with moderate intensity → no improvement in performance/VO₂max. DOI: 10.3389/fphys.2014.00034 .
APNM 2016 (Watsford et al.): investigated the effects of high- vs. low-cadence interval training on, among other things, freely chosen cadence and performance. DOI: 10.1139/apnm-2015-0562 .
➡️ Interim conclusion at the time: Low cadence can be effective – but not always. And: 40 rpm moderate seemed to be too "unspecific" or "not intense enough" for many (especially for highly trained athletes). Kristoffersen himself discusses this as well.
Phase 3: Mechanisms & Nuances (2016–2023)
Recent studies have increasingly shown that cadence influences not only "muscular vs. cardio" but also neuromuscular and even cortical activation (brain control of movement), which explains why some athletes experience greater "neurological" demands at very high cadence. This makes torque training interesting as a complementary stimulus – not as a "better" cadence.
Phase 4: The current study (2024) – Low Cadence within Polarized 🧨
Hebisz & Hebisz (2024, PLOS ONE) tested two 8-week polarized programs on well-trained female cyclists:
POL with freely selectable cadence (>80 rpm)
POL with low cadence (50-70 rpm) in SIT & HIIT
Result: The low-cadence program was more effective for key aerobic markers (including VO₂max, Pmax, VT2). DOI: 10.1371/journal.pone.0311833
Why this is important:
Previous studies failed partly due to (a) insufficient intensity, (b) very low cadence (40 rpm), or (c) incorrect embedding. Here, low cadence is deliberately integrated into high intensities – this is a different stimulus than "Z2 at 40 rpm".
3) What are the advantages of torque training? ✅
For cyclists (road/TT/gravel)
Muscular endurance at the pedal: particularly relevant for long climbs, headwinds, TT aero position.
Steady power at low cadence (realistic on hills/gravel) – this can be specific.
In a polarized context, low cadence at intervals can trigger additional adjustments (PLOS 2024).
For triathletes (sprint to long distance)
LD/Middle Distance: Torque training can help to remain economically "stable" at moderate performance levels – but only if it does not destroy running quality.
Sprint/Short: rather measured (short/targeted), otherwise "too much muscle" and too little speed.
For coaches: Why the appeal is plausible
Torque training increases pedal force demand at the same power output – this can (depending on the athlete) affect muscle recruitment and fatigue. This isn't automatically "better," but it is potentially beneficial.
4) What are the arguments AGAINST it? ⚠️
Knee/patellar tendon & hip : high torques can be problematic, especially with pre-existing strain.
"Spiral of death" in ERG mode : If cadence drops, the trainer increases resistance → even harder → cadence drops further. (Practical problem, not primarily a study problem.)
Non-specific if you make it too extreme (e.g. 40 rpm, moderate intensity) – see Kristoffersen 2014 (no effect in highly trained veterans).
5) Nutrition: What should you consider before/during Torque sessions? 🍌🥤
Torque/low-cadence intervals can feel more "expensive" for the muscles . Depending on the workout (tempo/threshold/SIT/HIIT), you often want quality – and carbohydrate availability is a key factor in achieving that.
Recommendation as a coach (pragmatic & evidence-based)
Before the session (during intense torque intervals): it's best not to "train low".
During (if >75–90 min or high density): Consume carbohydrates – this supports performance and reduces mental fatigue. A recent sports medicine review summarizes modern carbohydrate strategies for endurance (including dosage/application).
If you periodize carbohydrate restriction (“train low”), do so in a targeted manner and not at the expense of key stimuli; a meta-analysis of periodized carbohydrate restriction shows that performance effects are not guaranteed and depend heavily on the setting.
Practical rules:
SIT/HIIT low cadence: Carbs before and possibly during → ensure quality
Z2 Torque (short, 60–75 min): can also be done “low-ish”, but only if you really keep the session easy and don't need a second quality session the following day.
6) For whom (beginner/amateur/pro) and when in the annual plan? 🗓️
This is where most mistakes happen in practice: wrong timing + wrong dose.
Beginners 🟢
Goal: Technique + stress tolerance, not "K3 hero".
One block per week is sufficient: 6×3′ easy (Z2) @ 60–70 rpm
No ERG pressure, rather "resistance" and clean pedaling.
Stop when pulling on your knee/back.
Amateurs 🟠 (Job/Family)
Goal: specific strength endurance without ruining the week.
1 unit/week (3-6 week block), then reduce again.
Good choice: 3×10′ Tempo/Sweetspot @ 55–65 rpm (controlled).
Absolutely essential: Cooling + Fueling, otherwise you'll quickly experience fatigue.
Professionals 🔴
Goal: very specific (mountain, TT, gravel) or “Polarized plus”.
Short blocks (3-6 weeks), then more "normal cadence".
Low cadence in HIIT/SIT is a harsh neuromuscular stimulus – it doesn't fit into every race phase.
Age aspect 👴👩🦳
Masters athletes often benefit from strength endurance training, but are also more prone to tendon/knee problems.
The Kristoffersen finding (veterans, 40 rpm moderate) warns: being too extreme/too unspecific can be ineffective.
Gender ♀️♂️
The current PLOS study is interesting because it examines well-trained female athletes – an area where data is often lacking. My coaching interpretation: Torque blocks should be individualized for women (recovery, cycle/stress, injury history), not a "pink and shrink" approach.
7) Figure 2: Where does it fit into the annual plan + session toolkit
8) Transfer to running/trail: Does torque training on a bike make sense for runners? 🏃♂️➡️🚴♀️
Yes – but not in the way many hope.
Useful for runners/trail runners:
as a low-impact strength endurance stimulus , especially during high running loads or in injury-prone phases
to support overall fatigue resistance (durability)
as a substitute stimulus when downhill/impact is not possible
Limits:
Running performance is highly tendon-specific and elastic (SSC) – it cannot replace cycling.
Torque can help, but it's not running strength training .
Conversely: Does a long run benefit a cyclist? For health/robustness, yes – for pure cycling performance, only to a limited extent (and orthopedically riskier). In short: Use crossover training strategically, not blindly.
9) concrete torque workouts per goal 🎯
General guidelines (for all workouts):
Cadence ranges : typically 55–70 rpm (torque), not continuously below 50 rpm (except very selectively/briefly).
Resistance : choose a setting that allows you to step "smoothly" (no jerking, no "stomping").
Pain rule : If you experience a pulling sensation in your knee/patellar tendon, stop immediately or increase cadence/reduce power output.
Goal 1: Mountain / Gravel / “muscular endurance” (long sections of high pressure) ⛰️🪨
Workout A – Torque Tempo Block (Base/Build)
20' Roll-in (incl. 3×20'' spin-ups @ 100–110 rpm)
3×12' @ tempo to sweet spot (RPE 6/10), 60–65 rpm , 6' easy
10–15' Roll out
Why: long, controlled torque without being "too harsh". A very good entry point for amateurs.
Workout B – Over/Under Torque (Build)
15–20' Rolling in
2–3× (6' @ just below threshold 60–70 rpm + 2' @ sweet spot 55–60 rpm)
6' easy between the blocks
Why: simulated "ascent with edges" (alternating gradient/surface).
Workout C – Long Ride with Torque Insertions (Gravel-specific)
2.5–4h Z2
every 30-40 min: 8' @ Z2/tempo 60-65 rpm
Why: Durability + seating position + fueling under "muscular pressure".
Fueling tip:
For workouts A/B: 30–60 g carbohydrates/hour (depending on duration/intensity)
At C: 60–90 g KH/h (training for the race)
Goal 2: Time trial / Triathlon (steady power + aero tolerance) 🕒🚴♂️
Workout A – Aero Torque Sweet Spot
20' Roll in
2×20' @ 85-92% FTP, 65-75 rpm in Aero , 8' loose
Roll out
Why: steady pressure in aerodynamics, without "destroying" it.
Workout B – Threshold Torque in Aero (only if stable)
20' Roll in
3×10' @ 95-100% FTP, 65-70 rpm , 6' loose
Why: very specific for TT/70.3, but only if knees/back are okay.
Workout C – “Race Hold” without ERG (pacing transfer)
15–20' Rolling in
1×30–45' @ 80–90% FTP (race steady), 70–85 rpm , without ERG
Why: you learn to maintain performance on your own (simulating wind/reality).
Fueling tip:
From 60-75 minutes onwards, carbs and sodium are essential, especially indoors (sweat/heat!).
Goal 3: “Punch” / Road cycling (Accelerations, cornering, sprint preparation) ⚡
Torque can be useful here – but as a supplement , not as a main program.
Workout A – Short Torque Surges (neuromuscular)
20' Roll in
2× (6×30'' @ hard (RPE 8/10), 55–65 rpm + 90'' loose)
6 minutes easily between sets
Why: short, high-force impulse, without replacing sprinting technique.
Workout B – Torque → High Cadence Contrast
15–20' Rolling in
6× (2' @ tempo 60 rpm + 2' @ tempo 100 rpm) , 3' easy
Why: you train "force" and "speed" in one, very practical for everyday life.
Workout C – Sprint Prep (strength impulse + sprint)
20' Roll in
4× (45'' @ very heavy 55–60 rpm + 15'' “Sprint spin” 105–115 rpm) , 4' easy
Why: Activate Force, then rapid recruitment.
Fueling tip: With such units: better to be well fueled (carbs), otherwise quality suffers and you "die" early.
Knee-safe variant (age grouper, patellar tendon/knee sensitive) 🦵✅
Goal: Torque stimulus , but easier on the joints.
Knee-safe workout – “Moderate torque, higher cadence, less peak force”
15–20' Roll-in, 3×20'' Spin-ups
4×8' @ Z2/low tempo (RPE 4-5/10), 70-80 rpm , 4' easy
Optional (if pain-free): 4×20'' @ 60–65 rpm at the same power output (only a slightly harder gear), 2' loose
Roll out
Why this is knee-safe:
Cadence not extremely low → less brutal peak torque
Moderate intensity → less “pressure escalates”
You still get that specific "Force" touch
Additional rules for knee-safe progression:
Progression first over the entire time (e.g. 4×8 → 3×12)
only later at a slightly lower cadence
When in doubt , use more torque on the hill outdoors (briefly) rather than being "nailed" to the smart trainer.
When should I install it? (Quick reminder) 🗓️
Base/Build 1: ideal for a 3–6 week block (1x/week)
Build 2: more specific (e.g., TT-Aero Torque), but dose more carefully.
Competition phase: only "maintenance" (short/rare), focus on freshness
If, while reading this, you realize that you basically know what you want to train, but you're still missing the right dosage (cadence, intensity, progression, integration into your annual plan, knee/back compatibility, fueling) – then this is exactly the point where a quick look from the outside can often be very helpful.
I offer a virtual consultation (60 minutes) via MS Teams :
We will look at your goal, time budget, previous progress (and any knee/back issues) and use this to create a clear plan to determine if and how Torque training makes sense for you.
➡️ aecoaching.at (consultation) or office@aecoaching.at (The consultation is ideal if you don't want to "just do anything K3", but want to progress in a targeted and injury-free way.)
List of sources (selection, 20-year retrospective) 📚
Current core study
Hebisz R, Hebisz P. (2024). Greater improvement… low cadence (50-70 RPM) vs chosen cadence. PLOS ONE. DOI: 10.1371/journal.pone.0311833 .
Interventions & key older studies (for a 20-year perspective)
Nimmerichter A, et al. (2012). Low vs high cadence interval training… flat and uphill TT. Eur J Appl Physiol. DOI: 10.1007/s00421-011-1957-5 .
Kristoffersen M, et al. (2014). Low cadence (40 rpm) moderate intensity does not improve performance in veteran cyclists. Front Physiol. DOI: 10.3389/fphys.2014.00034 .
Watsford ML, et al. (2016). Low- vs high-cadence interval training…chosen cadence & performance. Appl Physiol Nutr Metab. DOI: 10.1139/apnm-2015-0562 .
Ludyga S, et al. (2016). High vs low cadence training and brain cortical activity. J Sci Med Sport.
Systematic review (imposed low cadence training): Effects of Cycling Training at Imposed Low Cadences – A Systematic Review.
Nutrition / Carbo-strategies
Valais GA. (2022). New Horizons in Carbohydrate Research and Application for Endurance Athletes. Sports medicine.
Systematic review/meta: Performance effects of periodized carbohydrate restriction in endurance athletes. JISSN (2021).
Comments