The 'Cycling Strength' Circuit Class Isn't Doing What You Think

The class is called something like "Cycle Strength" or "Power Endurance for Riders," and on a Wednesday evening I watched fourteen people in cycling shorts go through it in a glass-walled studio at a commercial gym. The coach had a stopwatch and a Bluetooth speaker. Twenty bodyweight squats. Twenty walking lunges. Twenty box jumps onto a low plyo box. Twenty kettlebell swings with a 12 kg bell. Twenty plank shoulder taps. Forty-five seconds of work, fifteen seconds of transition, two rounds. By the eight-minute mark the room smelled like a damp jersey and most of the heart rate straps on the wall display had crossed 170.

Everyone thought they had just done a strength session for cyclists. Most of them had paid extra for the class because their cycling coach told them they "needed to add strength work." Nobody was lying to anyone. And almost none of what happened in that room was going to make any of those riders measurably stronger or faster on the bike.

The Class Everyone Takes And Nobody Questions

The circuit-class formula has become the default version of "strength for cyclists" in commercial gyms, on YouTube, and inside cycling clubs running off-season group nights. Six to ten exercises in a circuit, fifteen to thirty reps per station, short rests, bodyweight or light kettlebells. The coaching cues are "keep your heart rate up" and "this is what your legs feel like in the last hour of a century ride." Everyone leaves soaked and convinced they have addressed the missing piece in their training.

The format makes sense if you ask the wrong question. The actual question — "what kind of strength training improves cycling performance" — has been studied for forty years, and the answer is not what the circuit class delivers.

In Dorsi we kept getting questions from cyclists about whether their circuit class counted as a strength session. The honest answer is no, but the question deserved a better explanation than that.

What The Cycling Literature Actually Says

The clearest single sentence on this comes from Rønnestad and Mujika's 2014 review in the Scandinavian Journal of Medicine and Science in Sports. They had spent a decade running and reviewing the relevant trials, and they put the conclusion in unusually direct language for a peer-reviewed paper:

"Cycling economy is improved by heavy strength training, but not by explosive or muscular endurance training in cyclists."

That is the entire debate in one sentence. The protocols that move the needle on cycling economy, time trial performance, and sprint power are heavy — at or above 80 percent of one-rep max, with sets of four to ten reps. The protocols that don't move the needle are the high-rep, explosive, circuit-style sessions that dominate group training rooms.

Aagaard and Andersen's 2010 review, in the same journal, walked through the mechanism. Endurance athletes get an enormous amount of metabolic stress from their endurance training. A trained cyclist riding fifteen hours a week has already saturated their slow-twitch fibers with metabolic signaling. What they are not getting from the bike is high motor-unit recruitment, high tendon strain, and the neural drive associated with maximal contractions. Those stimuli are only available from heavy resistance work. A 20-rep bodyweight squat doesn't produce them. A 20-rep set with a 12 kg kettlebell doesn't produce them. The threshold for recruiting the high-threshold motor units that drive adaptation sits somewhere north of 80 percent of one-rep max.

The most recent meta-analysis, Llanos-Lagos and colleagues in European Journal of Applied Physiology in 2025, pooled seventeen randomized controlled trials covering 262 cyclists. The "effective protocol" that emerges from the average across those trials is remarkably consistent: about 84 percent of 1RM, roughly a seven-rep max, about 3.5 sets per exercise, twice a week, for about thirteen weeks. The effects on cycling economy, sprint power, and time-trial output are real and replicable. Higher-rep, lower-intensity protocols in the same review did not produce the same effects. The signal was in the heavy work.

The circuit class is on the wrong side of every one of those numbers. Twenty reps with body weight is, generously, twenty percent of 1RM. The intent is to fatigue, not to overload. The two protocols are not different doses of the same medicine. They are different medicines.

Why The Circuit Class Doesn't Transfer

The cleanest way to understand why is to look at what cyclists are actually short on.

They are not short on metabolic stress. A trained rider doing two hours in zone 2 has produced more cumulative metabolic stress on their legs than any thirty-minute circuit class could. The glycolytic, slow-twitch endurance pathway is the one their sport is already specialized for. Adding more through a circuit is like watering a plant standing in a puddle.

They are short on mechanical stress. Cycling, mechanically, is a smooth, sub-maximal action with no eccentric loading and no impact. The peak forces in a pedal stroke are tiny compared to what a leg can produce in a heavy squat. The tendons never see high strain. The high-threshold motor units never get recruited because they are never needed. Over years of riding, this produces a paradoxical kind of weak fitness — enormous aerobic capacity sitting on top of a neural and tendinous system that has atrophied from disuse.

Heavy strength training repairs that gap. Aagaard and Andersen described three mechanisms in parallel. Neural drive improves — recruitment, rate coding, synchronization — so the existing muscle fires harder per cross-section. Tendon stiffness increases, so less energy is lost in the elastic phase of each pedal stroke. And the slow-twitch fibers themselves get strengthened enough that a steady 250 watts becomes a smaller percentage of their maximum, delaying fast-twitch recruitment and pushing the fatigue line further out.

None of those three mechanisms responds to a 20-rep kettlebell swing. They are all gated by load. Below the threshold of high-threshold motor unit recruitment, the body simply doesn't produce the adaptation. The circuit class delivers the input cyclists already have in excess and skips the input they actually lack.

What Actually Works Looks Boring

The honest replacement program is short, heavy, and aesthetically dull. Two sessions a week in the off-season, one session a week in-season, around 25 to 40 minutes per session. The lifts are unsurprising: back squat or front squat, Romanian or trap-bar deadlift, a single-leg pattern like a Bulgarian split squat, a hip hinge accessory like a hip thrust, and a small amount of trunk work. Five movements covering the mechanical demand of the pedal stroke.

The prescription is heavier than most cyclists expect. After three to four weeks of anatomical adaptation work at lighter loads, the main blocks run between three and six reps per set, three to five sets per movement, at loads of 80 percent of 1RM or higher. Rests are long — two to four minutes for the heavy compounds, ninety seconds for accessories. The session should not make you out of breath. If you are sweating through the squat block, the load is wrong, the rest is too short, or both.

In-season, Rønnestad and colleagues showed in 2010 that one session per week of two to three sets of four to six reps at around 80 percent of 1RM is enough to maintain off-season gains and continue to drive small improvements in 40-minute time trial output. One session. About thirty minutes. That's the dose.

The contrast with the circuit class is total. Circuit: high reps, low load, short rest, high heart rate, soaked t-shirts. Effective program: low reps, high load, long rest, low heart rate, no sweat to speak of. One of them looks like training. The other one is training.

A narrower exception worth naming: properly programmed plyometric work — loaded jump squats, trap-bar jumps, hang cleans — has a real place for sprinters and crit racers, usually as a brief power-transfer block at the end of a strength cycle. But the circuit class isn't that. Twenty-rep box jumps performed while already glycolytically smoked, with no recovery between reps, is conditioning with a hop in it. Real plyometrics requires maximal intent on every rep, which the high-rep, low-rest format makes impossible by definition.

The Short Version

If your strength session for cycling looks like a circuit — twenty-plus reps per exercise, short rests, heart rate above 150 the whole time — it is almost certainly not doing what you think it is doing. The literature has been consistent for fifteen years. The thing that improves cycling economy and time-trial performance is heavy resistance work at loads above 80 percent of 1RM, in sets of six or fewer, with long rests between sets. The thing that doesn't is everything else.

The circuit class isn't useless. It's a moderately fun group conditioning session that ends with everybody sweating. There's nothing wrong with that as a category. It's just not strength training for cyclists, and calling it that has led a generation of riders to spend their off-seasons not getting the adaptations they went to the gym to find.

You sweated. That's not nothing. It's just not the thing. Sweat is not the currency of adaptation; load is. The next four weeks, with the circuit replaced by a real strength session twice a week, will tell you which one was actually doing the work all along.