Flywheel Training: Is Eccentric Overload Worth It?

A flywheel has no weight stack and no plates. It stores the energy you put into it on the way up and hands it straight back on the way down, so the harder you pull, the harder you have to brake. That simple property is why flywheel (isoinertial) training has moved from European Space Agency bed-rest studies into elite football clubs, rehab clinics and, increasingly, commercial gyms.

The marketing claim is "eccentric overload": more load in the lowering phase than gravity-based training can give you. The question for a serious lifter or coach is whether that overload actually happens, and whether it builds more strength, size or power than a barbell does.

Why this matters

Eccentric work is the phase of a rep where the muscle lengthens under load. Muscles can produce substantially more force eccentrically than concentrically, so with a barbell the lowering phase is almost always underloaded relative to what the muscle could tolerate. Methods that close that gap (heavy negatives, weight releasers, two-up one-down) are awkward to set up and often need a spotter. A flywheel promises the same effect, rep after rep, from one compact device. If the evidence holds up, that is a genuinely useful tool. If it does not, it is an expensive way to do a squat.

How a flywheel actually loads the muscle

On a flywheel device, a strap is wound around a shaft attached to a spinning disc. During the concentric phase you unwind the strap and accelerate the disc. At the end of the range the strap rewinds and the stored kinetic energy pulls you back into the eccentric phase, which you must decelerate. Resistance is set by the moment of inertia of the disc (measured in kg·m²), not by a mass in kilograms. An international consensus statement from 19 flywheel researchers now recommends describing load by inertia and monitoring output by velocity or power, rather than borrowing barbell language.¹

Two features follow from this design. First, resistance is accommodating: every rep is as hard as your effort makes it, so a maximal concentric produces a maximal returning load. Second, overload is not automatic. You only get more eccentric than concentric output if you delay braking to the final part of the lowering phase, which concentrates the deceleration into a shorter window. In flywheel squats, heavier inertias reduced both concentric and eccentric velocity in a highly predictable, individual way (median R² about 0.95 for the inertia-velocity line), while the ratio of peak eccentric to peak concentric power rose modestly as inertia increased.² The authors concluded that velocity, not power, is the better variable for prescribing and tracking flywheel intensity.²

What EMG shows during flywheel reps

The clearest early evidence came from Norrbrand and colleagues, who trained knee extensors on either a weight stack or a flywheel for five weeks. With EMG normalised to maximal voluntary contraction, the flywheel group showed greater eccentric activation of vastus lateralis and medialis across joint angles, and near-maximal concentric activation from the very first session.³ The weight-stack group only reached near-maximal concentric activation after training. In other words, the flywheel demanded high neural drive in both phases from day one.

The picture is more nuanced in the squat. Comparing a flywheel squat with a 10-repetition-maximum barbell squat in strength-trained men, normalised quadriceps EMG did not differ between modes, yet MRI showed greater exercise-induced T2 signal change after the flywheel squat, particularly in rectus femoris (+24% versus +8%).⁴ Muscle use, judged by imaging, was at least comparable and arguably greater, even though surface EMG amplitude could not separate the two.

Fatigue tracking also behaves differently on a flywheel. During fatiguing concentric-eccentric flywheel actions, muscle fibre conduction velocity fell faster than the EMG mean frequency, so spectral measures underestimated the change that was occurring.⁵ For anyone interpreting median-frequency or spectral fatigue curves during dynamic, variable-force work, this is an important caution: the spectral signal is informative, but it is not a direct read-out of fibre physiology.

Strength, size and power: flywheel versus weights

The meta-analyses disagree in ways that are themselves instructive. An early systematic review of randomised trials pooled 9 studies (276 participants) and reported a moderate overall effect (SMD 0.63) favouring flywheel eccentric-overload training over conventional resistance training for strength, power, hypertrophy, jump height and running speed.⁶ A broader meta-analysis of 20 studies found that 4 to 24 weeks of flywheel training produced meaningful within-group gains in muscle cross-sectional area (effect size 0.59), maximal strength (1.33) and power (1.19), with the strongest evidence in trained younger athletes using short, intensive blocks.⁷

A stricter review, limited to studies that directly compared flywheel with gravity-dependent training, reached a different conclusion. Across three randomised and four non-randomised trials, flywheel training was not superior for muscle strength (SMD -0.05 in the randomised trials, I² = 0%).⁸ The most recent synthesis, with 16 studies and 341 participants, landed in between: no significant difference for maximal strength (SMD 0.24, confidence interval crossing zero), but a small advantage for muscle power (SMD 0.47).⁹ That power advantage held against weight-stack machines but was not significant against free weights.⁹

The hypertrophy question is similarly mixed. A meta-analysis of 13 studies found that flywheel training increased muscle volume, but the gain in muscle mass was not influenced by the presence of eccentric overload, whereas strength gains were.¹⁰ The best recent head-to-head trial, a volume-matched 10-week total-body programme in young adults, found that flywheel and traditional training produced similar increases in isometric strength and fat-free mass.¹¹ The difference was where and how muscle changed: flywheel training increased fascicle length and cross-sectional area in the distal vastus lateralis and improved jump performance, while traditional training increased pennation angle and proximal vastus lateralis size and produced larger barbell 1RM gains.¹¹ Each group got better at the task it trained.

The honest summary is that a flywheel is not a hypertrophy shortcut. It appears to be at least as effective as conventional loading for size and strength, modestly better for power, and different in the architectural signature it leaves behind.

Hamstrings, fascicle length and injury risk

Fascicle length matters because longer fascicles in biceps femoris long head are associated with lower hamstring strain risk, and eccentric training is the main known way to lengthen them. In a randomised trial, six weeks of flywheel leg curls performed with a deliberate eccentric bias increased biceps femoris long head fascicle length by 14%, while conventional flywheel leg curls produced no change.¹² The gain was lost after four weeks of detraining, and neither group improved eccentric or isometric strength or rate of force development.¹² The overload has to be coached, not assumed.

A preregistered within-subject trial reinforced this specificity. Six weeks of unilateral flywheel hip-extension and leg-curl training improved flywheel eccentric peak power substantially, but did not significantly improve isokinetic eccentric or isometric knee-flexor strength.¹³ Gains on the device did not fully transfer to other ways of testing the same muscle.

The most cited injury result is older but striking. In elite Swedish footballers, adding 10 weeks of preseason eccentric-overload hamstring training reduced hamstring strain injuries over the following season from 10 of 15 players in the control group to 3 of 15 in the training group, alongside improvements in strength and sprint speed.¹⁴ The sample was small, but the effect size explains why flywheels became common in football. In field-sport athletes more broadly, flywheel training also outperformed traditional training for change-of-direction performance (SMD 0.64), with the largest effects from two or fewer sessions per week and twelve or fewer sessions in total.¹⁵

What this means in practice

Flywheel training is best treated as a specialised eccentric and power tool rather than a replacement for barbell work. The consensus guidance points to a few practical principles: familiarise athletes over several sessions before judging output, select inertia by the velocity it produces, use the first one or two reps to spin the disc up rather than counting them, and coach a late, aggressive brake if eccentric overload is the goal.¹ Short blocks of six weeks or so, two sessions a week, are where most of the positive effects have been reported.⁷,⁹,¹⁵

This is also where a muscle-worn wearable earns its place. Because flywheel resistance is entirely effort-driven, two sets at the same inertia can be very different sets. Surface EMG on the target muscle can show whether drive in the eccentric phase is genuinely rising relative to the concentric, and IMU data can show the timing and velocity of the brake, which is the variable that determines whether overload happened at all.²,³ ZELOS interprets those signals as within-person comparisons: this set against your previous sets, this inertia against the last block. That framing matters, because the research is clear that EMG amplitude alone is not a growth dial.⁴,⁵ What it can tell you, in real time, is whether you are actually doing the kind of rep the method depends on.

Key takeaways

• Flywheel resistance is set by inertia and is accommodating, so eccentric overload only occurs when the lifter delays braking to late in the lowering phase.

• Flywheel exercise produces high eccentric and concentric neural drive, but surface EMG and spectral fatigue measures do not capture everything happening in the muscle.

• For maximal strength and hypertrophy, flywheel training is broadly equivalent to conventional resistance training; its clearest advantage is a small-to-moderate gain in muscle power.

• Eccentric-biased flywheel hamstring work can lengthen biceps femoris fascicles and has been linked to fewer hamstring strains in footballers, but the adaptation fades quickly without ongoing training.

• Prescribe by velocity, coach the brake, and use short, focused blocks.

References

1. Beato, M., de Keijzer, K. L., Muñoz-Lopez, A., Raya-González, J., Pozzo, M., Alkner, B. A., Dello Iacono, A., Vicens-Bordas, J., Coratella, G., Maroto-Izquierdo, S., Gonzalo-Skok, O., McErlain-Naylor, S. A., Martin-Rivera, F., Hernandez-Davo, J. L., Arrones, L. S., Sabido, R., de Hoyo, M., Fernandez-Gonzalo, R., & Norrbrand, L. (2024). Current guidelines for the implementation of flywheel resistance training technology in sports: A consensus statement. Sports Medicine, 54(3), 541–556. https://doi.org/10.1007/s40279-023-01979-x

2. McErlain-Naylor, S. A., & Beato, M. (2021). Concentric and eccentric inertia-velocity and inertia-power relationships in the flywheel squat. Journal of Sports Sciences, 39(10), 1136–1143. https://doi.org/10.1080/02640414.2020.1860472

3. Norrbrand, L., Pozzo, M., & Tesch, P. A. (2010). Flywheel resistance training calls for greater eccentric muscle activation than weight training. European Journal of Applied Physiology, 110(5), 997–1005. https://doi.org/10.1007/s00421-010-1575-7

4. Norrbrand, L., Tous-Fajardo, J., Vargas, R., & Tesch, P. A. (2011). Quadriceps muscle use in the flywheel and barbell squat. Aviation, Space, and Environmental Medicine, 82(1), 13–19. https://doi.org/10.3357/asem.2867.2011

5. Pozzo, M., Alkner, B., Norrbrand, L., Farina, D., & Tesch, P. A. (2006). Muscle-fiber conduction velocity during concentric and eccentric actions on a flywheel exercise device. Muscle & Nerve, 34(2), 169–177. https://doi.org/10.1002/mus.20574

6. Maroto-Izquierdo, S., García-López, D., Fernandez-Gonzalo, R., Moreira, O. C., González-Gallego, J., & de Paz, J. A. (2017). Skeletal muscle functional and structural adaptations after eccentric overload flywheel resistance training: A systematic review and meta-analysis. Journal of Science and Medicine in Sport, 20(10), 943–951. https://doi.org/10.1016/j.jsams.2017.03.004

7. Petré, H., Wernstål, F., & Mattsson, C. M. (2018). Effects of flywheel training on strength-related variables: A meta-analysis. Sports Medicine - Open, 4(1), 55. https://doi.org/10.1186/s40798-018-0169-5

8. Vicens-Bordas, J., Esteve, E., Fort-Vanmeerhaeghe, A., Bandholm, T., & Thorborg, K. (2018). Is inertial flywheel resistance training superior to gravity-dependent resistance training in improving muscle strength? A systematic review with meta-analyses. Journal of Science and Medicine in Sport, 21(1), 75–83. https://doi.org/10.1016/j.jsams.2017.10.006

9. Hu, Z., Liu, Y., Huang, K., Huang, H., Li, F., & Yuan, X. (2024). Comparing the effect of isoinertial flywheel training and traditional resistance training on maximal strength and muscle power in healthy people: A systematic review and meta-analysis. Life, 14(7), 908. https://doi.org/10.3390/life14070908

10. Nuñez Sanchez, F. J., & Sáez de Villarreal, E. (2017). Does flywheel paradigm training improve muscle volume and force? A meta-analysis. Journal of Strength and Conditioning Research, 31(11), 3177–3186. https://doi.org/10.1519/JSC.0000000000002095

11. Banks, N. F., Berry, A. C., Rogers, E. M., & Jenkins, N. D. M. (2024). Flywheel resistance training promotes unique muscle architectural and performance-related adaptations in young adults. European Journal of Sport Science, 24(12), 1765–1778. https://doi.org/10.1002/ejsc.12215

12. Presland, J. D., Opar, D. A., Williams, M. D., Hickey, J. T., Maniar, N., Lee Dow, C., Bourne, M. N., & Timmins, R. G. (2020). Hamstring strength and architectural adaptations following inertial flywheel resistance training. Journal of Science and Medicine in Sport, 23(11), 1093–1099. https://doi.org/10.1016/j.jsams.2020.04.007

13. de Keijzer, K. L., McErlain-Naylor, S. A., & Beato, M. (2024). Six weeks of unilateral flywheel hip-extension and leg-curl training improves flywheel eccentric peak power but does not enhance hamstring isokinetic or isometric strength. International Journal of Sports Physiology and Performance, 19(1), 34–43. https://doi.org/10.1123/ijspp.2023-0035

14. Askling, C., Karlsson, J., & Thorstensson, A. (2003). Hamstring injury occurrence in elite soccer players after preseason strength training with eccentric overload. Scandinavian Journal of Medicine & Science in Sports, 13(4), 244–250. https://doi.org/10.1034/j.1600-0838.2003.00312.x

15. Chaabene, H., Markov, A., Prieske, O., Moran, J., Behrens, M., Negra, Y., Ramirez-Campillo, R., Koch, U., & Mkaouer, B. (2022). Effect of flywheel versus traditional resistance training on change of direction performance in male athletes: A systematic review with meta-analysis. International Journal of Environmental Research and Public Health, 19(12), 7061. https://doi.org/10.3390/ijerph19127061