Skip to content

Strength, warm-ups, and mobility

Last checked: 2026-10-02

Scope: Questions 18 and 19 only. Heavy, maximal, explosive, plyometric, core, and low-load “activation” training, and their effects on cycling economy, sprint power, time-trial or time-to-exhaustion performance, and injury resilience. Concurrent-training interference, session spacing, delayed-onset muscle soreness, progression, and maintenance. Acute warm-ups, glute and band routines, dynamic and static stretching, and pre-ride mobility. Audience: a trained amateur with about 7–12 hours per week available in total, including strength, with a note for the case where those hours were counted as cycling only. Transfer notes cover a 20–30 hour professional week.

Platform navigation: Reference contents · CONTEXT · bibliography

Reading order: Previous — sleep, heat and iron · Next — load and adaptation monitoring.

This file is a living topic brief. It is not a clinical protocol. Pain that changes pedalling, joint swelling, neurological symptoms, or crash injury belongs with a clinician and a bike fit, not with a generic activation circuit.

How this brief was checked

No transcript text is in initial_prompt.md. The prompt only paraphrases that the transcript contrasts low-load activation with genuine strength training. That paraphrase is a hypothesis to test, not a confirmed quotation and not a paper attribution.

Literature was retrieved from Europe PMC (including full text where a PMCID was available), PubMed abstracts, and publisher pages. web_search was not available at the start of this investigation and was used after access was restored for 2025–2026 discovery and cross-checks. Search-engine summaries were not used as the sole source of effect sizes, sample sizes, or confidence intervals. Where an abstract’s markup dropped a number, that number is omitted rather than reconstructed. Publications with a first date later than 2026-10-02 were not used.

The latest cyclist-specific synthesis is one paper with two dates: Llanos-Lagos, Ramirez-Campillo, and Sáez de Villarreal, online 9 July 2025, then the 2026 journal issue of European Journal of Applied Physiology, volume 126, issue 1, pages 193–222 (doi:10.1007/s00421-025-05883-2; PMC12881108). Calling it either “the 2025 review” or “the 2026 review” without the other date is ambiguous. Searches indexed in that review ran to February 2025. A title search through 2026-10-02 did not find a newer cyclist strength meta-analysis. It did find on-bike versus gym trials, a quasi-experimental FIFA 11+ warm-up study, and practice surveys.

Question navigation

Durable conclusions

STR-C01 — Heavy lower-body strength can improve cycling performance without raising VO₂max

Classification: Moderate evidence

Claim: Adding heavy lower-body strength work (≥80% of 1RM in the 2025/2026 meta-analysis; typically 4–10RM, twice weekly, for about 8–25 weeks in the primary trials) to ongoing cycling improves time-trial or time-to-exhaustion performance in trained cyclists by a small-to-moderate pooled amount, and does not reliably increase VO₂max.

Support: Llanos-Lagos et al. pooled 17 controlled comparisons, 262 cyclists (60 women), mean baseline VO₂max 61.25 ml·kg⁻¹·min⁻¹. Versus cycling without heavy strength, performance (time trial and time to exhaustion combined) had Hedges’ g 0.463 (95% CI 0.109 to 0.817, p = 0.016). VO₂max was null (g −0.041, 95% CI −0.345 to 0.263, p = 0.773), as were power at VO₂max and maximal metabolic steady state. Certainty was low, mainly from moderate risk of bias and imprecision, including non-random allocation in some included studies. Primary trials in the same direction include Rønnestad, Hansen, and Raastad (2010; 2011), Sunde et al. (2010), Aagaard et al. (2011, 45-minute capacity), Vikmoen et al. (2016), and Sitko, López-Laval, and Cirer-Sastre (2024).

Contradiction: Not every trial moves the endurance test. Bishop et al. (1999) increased squat strength in endurance-trained women and concluded that this did not improve cycle endurance performance. Jackson, Hickey, and Reiser (2007) found no added benefit on a lactate-profile test or economy after 10 weeks in club cyclists. Rønnestad, Hansen, and Nygaard (2017) reported a 40-minute all-out change of 4 ± 6% versus −1 ± 6% after 10 weeks in elite cyclists (P = 0.13).

Practical implication: Use heavy strength as an extra stimulus for sustainable power and short power, not as a substitute for VO₂ or threshold riding. Do not expect the gym to raise the VO₂max ceiling.

Uncertainty: The pooled performance effect is real in the statistical sense and small-to-moderate, but GRADE certainty is low. Optimal sets, exercises, and weeks are not identified: participant and programme moderators were non-significant. Several positive trials come from one research group and some share a cohort.

STR-C02 — Economy and anaerobic power are plausible mediators; threshold is not a guaranteed adaptation

Classification: Moderate evidence

Claim: When heavy strength improves cycling performance, the better-supported companions are a small improvement in cycling efficiency and a moderate improvement in anaerobic power (peak sprint power). A higher lactate or ventilatory threshold is not a reliable group effect. Strong local mechanisms — later recruitment of less efficient fibres, type IIX toward IIA, tendon stiffness, earlier peak torque — do not validate one exact set-and-rep prescription.

Support: In Llanos-Lagos et al., cycling efficiency showed a small beneficial effect of magnitude about |g| 0.353 (p = 0.012), with low certainty. The prose states a positive point estimate of 0.353 together with a negative 95% CI (−0.619 to −0.088). Figure 3 displays a negative pooled estimate, and its caption says a negative effect size is beneficial. That positive point plus negative interval is a source reporting inconsistency, not a valid signed pair, and it is not corrected here into a new estimate. Anaerobic power g was 0.560 (95% CI 0.097 to 1.023, p = 0.024). Anaerobic capacity (mean sprint power) and maximal metabolic steady state were not significant. Performance g 0.463 (0.109 to 0.817) is unchanged and is a coherent signed result. Sunde et al. (2010) reported +4.8% cycling economy and +17.2% time to exhaustion at the pre-intervention maximal aerobic power after 8 weeks of 4 × 4RM half-squats, with no VO₂max change. Vikmoen et al. (2016) reported better 40-minute power, fractional utilisation, and economy in well-trained women, with a shift from IIAX–IIX toward IIA.

Contradiction: Rønnestad et al. (2015) improved 40-minute power, peak aerobic power, and power at 4 mmol·L⁻¹ lactate in young elite cyclists over 25 weeks, with earlier peak torque, but neither VO₂max nor cycling economy changed. Aagaard et al. (2011) found economy improved in the endurance-only young elites until it matched the strength group. Individual threshold gains in some Rønnestad trials therefore do not overturn the pooled null for maximal metabolic steady state.

Practical implication: Judge the block by whether short power and a race-relevant effort improve, not by whether VO₂max or FTP “must” move. Strength does not own a separate energy system that intervals cannot also train.

Uncertainty: The meta-analysis did not test mediation. Efficiency protocols differ (gross, delta, economy at different powers), which is why a three-level model was required. Effect-size thresholds in that paper were 0.15, 0.45, and 0.80 for small, moderate, and large. The efficiency magnitude is usable; the printed positive point with a negative interval is not a coherent confidence interval.

STR-C03 — Sprint and plyometric work are specific; they are not a proven replacement for heavy strength

Classification: Moderate evidence for task specificity; Plausible but uncertain that plyometrics alone improve cycling economy or time-trial performance

Claim: Short on-bike sprints improve sprint power more than continued heavy strength once a strength base exists. Isolated plyometrics have not been shown, in a cyclist trial retrieved here, to improve economy or time-trial performance on their own. A combined jump-plus-low-cadence-sprint protocol can raise short time-trial power, but the cycling sprints confound the jumps.

Support: Kristoffersen et al. (2019) randomised 28 competitive cyclists (VO₂max 61.1 ± 5.9 ml·kg⁻¹·min⁻¹), after 4 shared weeks of strength, to 6 weeks of heavy strength or short-sprint training. Short sprints were superior for 6-second, 30-second, and 55 m bike sprints. Heavy strength produced more maximal strength. Endurance-capacity changes did not differ between groups. Paton and Hopkins (2005) replaced part of usual training for 4–5 weeks with 12 sessions of single-leg jumps plus 5 × 30-second high-resistance cycling sprints. Versus control, 1 km power changed by about +8.7% and 4 km power by about +8.1% (90% confidence limits reported in the abstract). That is not a plyometric-only trial. Kordi et al. (2020) improved peak power by about 3–4% in elite track sprinters with 6 weeks of cycling-specific isometric strength; that population is not a road amateur.

Contradiction: Rønnestad and Mujika (2014), a narrative review, recommended heavy rather than explosive strength for cycling economy, while either mode can improve running economy. Berryman et al. (2018) pooled middle- and long-distance sports, not cyclists alone, and found larger performance effects from maximal-force training than from other strength intensities (overall performance SMD 0.52, 95% CI 0.33 to 0.70). Beattie et al. (2017) combined maximal and explosive work for 20 weeks in 6 versus 9 competitive road cyclists and reported gains in maximal strength and bike-specific explosive strength; the retrieved abstract does not cleanly report the economy result.

Practical implication: If the limiter is the 5–15 second effort, practise that effort on the bike. If the limiter is repeatable power over minutes, keep heavy strength or an equally high-force on-bike substitute (see below). Do not add a long plyometric circuit “for economy” without a performance reason.

Uncertainty: Head-to-head explosive-only versus heavy-only trials in trained road cyclists remain scarce. Paton’s confidence limits for lactate-profile power crossed a trivial change.

STR-C04 — Core drills and low-load activation are not a required performance method

Classification: Moderate evidence that conventional heavy strength beats this low-load core programme for the power profile; Mostly coaching practice that glute activation, bands, or core circuits are required before riding. There is no adequate evidence that these routines correct a claimed imbalance or prevent cycling injury. That is an evidence gap, not a proof that every such routine is ineffective.

Claim: A 12-week core routine (glute bridge, plank, prone back extension, twice weekly) did not match heavy strength for 5-second through 20-minute power in trained road cyclists. Electromyography during a band or therapeutic exercise shows that a muscle can be activated. It does not show that the exercise improves time-trial power or economy. There is no adequate evidence that these routines correct a claimed imbalance or prevent cycling injury.

Support: Sitko et al. (2024) divided 36 trained road cyclists into cycling only, cycling plus core, or cycling plus conventional strength (half-squat, single-leg leg press, single-leg hip flexion, ankle plantar flexion; 6RM to failure; twice weekly; 12 weeks). Adjusted 20-minute relative power rose about 0.22 W·kg⁻¹ with strength versus about 0.07 and 0.06 W·kg⁻¹ in cycling-only and core. Strength was also superior at 5 seconds, 60 seconds, and 5 minutes. Core beat cycling-only only at 5 seconds, and the cycling-only group lost sprint power. Distefano et al. (2009) quantified gluteal EMG during therapeutic exercises in 21 healthy adults. That is a recruitment study. Crow et al. (2012) found that low-load glute exercises acutely raised countermovement-jump peak power in 22 elite Australian footballers versus no warm-up. Those athletes were not cyclists, and the test was not a time trial.

Contradiction: Abt et al. (2007) fatigued the core of 15 competitive cyclists and then observed more frontal-plane knee motion (about 15° to 23°) during treadmill cycling, without a change in pedal force. The authors inferred injury risk. That is an acute fatigue model, not a training trial and not an injury count. Coaching programmes that prescribe daily band walks are not contradicted by a cyclist RCT because that RCT has not been retrieved; absence of benefit is not the same as proof of harm. The cost is time and false precision.

Practical implication: If strength time is scarce, spend it on heavy lower-body work or truly high-force pedalling, not on a long activation sequence. A few easy pedal revolutions are enough to “switch on” the pedalling muscles before an easy ride.

Uncertainty: Sitko estimated VO₂max from 5-minute power, so the reported VO₂max rise is not a gas-exchange finding and should not be stacked on top of the power result. Participants were not blinded. One small core-versus-strength trial is not every possible trunk programme.

STR-C05 — Interference is real, but a universal 6-hour rule is not a cyclist result

Classification: Moderate evidence that concurrent endurance can blunt explosive strength and that muscle damage can blunt the next hard cycling effort; Plausible but uncertain that 6 hours is the cyclist optimum versus a separate day

Claim: The classic interference question is usually “does endurance blunt strength?”, not “does strength blunt VO₂max?”. Cycling is generally a less interfering endurance mode than running for hypertrophy and maximal strength. Explosive-strength gains are more fragile, especially when both modes are in the same session. Spacing of 6 hours versus 24 hours has been tested in amateur rugby, not in cyclists. Residual muscle damage can still make the next day’s sprint and incremental peak power worse even when a pooled time trial does not move.

Support: Schumann et al. (2022) pooled 43 studies of concurrent aerobic plus strength training versus strength alone. Maximal strength (SMD −0.06, 95% CI −0.20 to 0.09) and hypertrophy (SMD −0.01) were not compromised. Explosive strength was (SMD −0.28, 95% CI −0.48 to −0.08), more clearly when modes were in the same session than when separated by at least 3 hours. Cycling versus running did not moderate those outcomes in that update. Wilson et al. (2012), an earlier meta-analysis, did find decrements in hypertrophy and strength when the endurance mode was running rather than cycling, and negative correlations with endurance frequency and duration. Rønnestad, Hansen, and Raastad (2012) showed the practical cyclist version: well-trained cyclists doing the same heavy strength programme as non-endurance trainees gained less 1RM, thigh size, jump performance, and rate of force development. Robineau et al. (2016) assigned 58 amateur rugby players to 0, 6, or 24 hours between strength and aerobic work for 7 weeks. Zero hours was worse for maximal strength; 24 hours favoured some high-velocity strength and the VO₂peak gain. Held et al. (2026), an umbrella review of 17 meta-analyses (search to 28 February 2025; recreationally trained emphasis), found concurrent training comparable to resistance training for strength, power, and hypertrophy, and better than endurance-only for strength. Highly trained data were scarce. The retrieved abstract did not establish a 6-hour versus 24-hour rule.

Contradiction: Eddens, van Someren, and Howatson (2018) found a modest advantage for doing resistance before endurance in the same session for lower-body dynamic strength (about 6.9% difference) and no sequence effect on hypertrophy or VO₂max. Murlasits, Kneffel, and Thalib (2018) found about 4 kg more lower-body 1RM when strength came first, and no sequence effect on aerobic capacity. Sequence is a small lever. It is not a reason to build an elaborate same-day puzzle. Devantier-Thomas et al. (2024) found that exercise-induced muscle damage reduced sprint and incremental peak power and raised ventilation and perceived exertion in submaximal cycling at 24–48 hours, while pooled time-trial and distance-trial performance did not change significantly. “No significant time-trial change” is not proof that a key session is unaffected.

Practical implication: Put heavy strength on its own day when you can. If it must share a day with riding, do not place a new or very sore heavy session immediately before the week’s hardest intervals. Six hours is a reasonable minimum borrowed from non-cyclists, not a tested cycling optimum. A separate day is the cleaner hedge. Do the quality ride first if both must be close and the ride is the priority; do strength first only if the strength adaptation is the priority and the ride is easy.

Uncertainty: Almost all spacing meta-analyses enrol non-cyclists. Transfer to a 7–12 hour cyclist is plausible because cycling interferes less than running, but the exact gap has not been trialled against cycling performance.

STR-C06 — Two sessions a week can build the effect; one often maintains it; the minutes and the recovery cost both count

Classification: Moderate evidence for the build-and-maintain pattern in small cyclist trials; Mostly coaching practice for the minute-by-minute exercise menu, which is a synthesis of those trials rather than a single optimised protocol

Claim: The interventions that improved cycling performance usually used two lower-body strength sessions per week for at least 8–12 weeks. One session per week can maintain size and strength once it has been built. That pattern is a supported implementation, not proof that 30 minutes is the optimal recipe. Session length in the successful trials was often on the order of 30–45 minutes, and Llanos-Lagos et al. did not identify an optimal duration. The minutes and the recovery cost must be counted. If 7, 9, or 12 hours means cycling only, lifting adds about 1–1.5 hours in a two-session build, or about 30–45 minutes when maintaining once weekly. If those hours are total available time, strength replaces easy cycling minutes rather than sitting outside the budget. Hiding the work inside a ride file still understates load.

Support: Rønnestad et al. (2010, European Journal of Applied Physiology) used four lower-body exercises, 3 × 4–10RM, twice weekly for 12 weeks. A companion maintenance study in six cyclists kept thigh size and strength with one heavy session per week for 13 competition weeks and continued to improve performance determinants relative to endurance-only training. Vikestad, Lyngstad, and Dalen (2025) found that professional road cyclists commonly use two strength sessions in the off- and pre-season and drop to one in the race season. That is practice, not proof, but it matches the trial pattern. Llanos-Lagos et al. could not show that frequency, duration, or strength experience moderated the pooled effect, and they explicitly declined a robust optimal prescription. Included programmes often used multi-joint and single-leg work for 3–4 sets.

Contradiction: Yamamoto et al. (2010) reviewed five small trials to September 2009 and concluded that replacing some endurance training with resistance work was more often successful than adding resistance on top. Later add-on trials (Rønnestad, Sunde, Vikmoen, Sitko) weaken that as a universal rule. The time-budget point survives: adding strength without protecting recovery can still spoil the cycling. Pallarés et al. (2025) and Barranco-Gil et al. (2025) show that stopping a strength stimulus while continuing about 11 hours of cycling was associated with loss of strength and quadriceps size. Those papers should be read as one research programme, not two independent replications.

Practical implication: In the integrated weeks, total availability is split rather than exceeded: about 390 minutes of cycling plus 30 minutes of strength at 7 hours, 480 plus 60 at 9 hours, and 660 plus 60 at 12 hours. A two-session build needs either a larger swap of easy cycling or, if the quoted hours were cycling-only, an extra 1–1.5 hours. Do not treat 30 minutes as a magic dose.

Uncertainty: Maintenance evidence is one small cyclist study (n = 6 per arm) plus professional self-report. The minimum effective dose for an already strong amateur is not known.

STR-C07 — A short cycling warm-up changes output; a long activation routine does not have to

Classification: Strong evidence that static stretches of about 60 seconds or longer per muscle can acutely reduce isolated force, and that shorter stretches usually do not; Moderate evidence that a cycling warm-up improves short time-trial or severe-intensity work versus no warm-up; Mostly coaching practice that bands, glute drills, or a fixed mobility flow are required

Claim: For a hard effort lasting a few minutes to about 12 minutes, the useful warm-up is easy-to-moderate cycling plus a few openers, long enough to raise muscle temperature and speed oxygen-uptake kinetics, and short enough not to leave residual fatigue. Prolonged static stretching is the component with a real acute cost to isolated strength. Dynamic movement and very short static stretches are optional. They are not a performance prerequisite.

Support: Burnley, Doust, and Jones (2005) found that prior moderate or heavy cycling raised mean power in a 7-minute performance trial by about 2–3% in 12 well-trained cyclists (control 330 ± 42 W; moderate 338 ± 39 W; heavy 339 ± 42 W). A sprint prime did not help (324 ± 45 W). Hajoglou et al. (2005) concluded that either an easy or a hard warm-up improved 3 km time-trial performance versus no warm-up in 8 well-trained road cyclists; the exact wattage differences were not recoverable from the retrieved abstract. Tomaras and MacIntosh (2011) showed the other error: a traditional track warm-up of at least 50 minutes with repeated sprints reduced 30-second peak power (1303 ± 89 W versus 1390 ± 80 W after a shorter, easier warm-up). Jaén-Carrillo et al. (2026; first published 22 September 2026) found that 45 minutes of 42 °C lower-limb hot-water immersion matched a standardised active warm-up for 12-minute time-trial power in 27 tier 2–3 cyclists training 9 ± 3 h·week⁻¹ (259 ± 52 W versus 255 ± 53 W, P = 0.308). Temperature is a large part of the effect. A 45-minute bath is not a practical upgrade. Kay and Blazevich (2012) limited the static-stretch force deficit mainly to holds of at least 60 seconds. Behm et al. (2016) reported about −4.6% performance with ≥60 seconds per muscle versus about −1.1% with less than 60 seconds, and no clear injury-prevention effect of static or PNF stretching. Warneke and Lohmann (2024) revisited controlled pre–post trials (83 studies, 2012 participants): isolated maximal strength still fell (g −0.21 overall; −0.84 when a bout was ≥60 seconds versus passive control), but athletic performance in general did not, and adult jumping was slightly positive (g 0.15). The review is not restricted to trained cyclists.

Contradiction: Silva et al. (2014) found that 4 × 5RM leg press 10 minutes before a 20 km time trial cut time by 6.1% in 11 trained cyclists. That is acute post-activation performance enhancement, not a mobility routine, and it is not a reason to lift heavy before every race. Blazevich and Babault (2019) separate true twitch potentiation, which decays in seconds, from voluntary performance changes minutes later. Responses are individual. Chaabene et al. (2019) argue that ≤60 seconds of static stretching inside a full warm-up only trivially impairs strength and power (about 1–2%), whereas longer holds cost about 4–7.5%, and they suggest a possible injury benefit in sprint and change-of-direction tasks. That injury suggestion conflicts with Behm et al. (2016), Small, Mc Naughton, and Matthews (2008), and Lauersen, Bertelsen, and Andersen (2014), who did not find stretching effective for injury prevention. Prefer the injury trials over the mechanistic suggestion.

Practical implication: Before a quality session or race, ride. Use 10–20 minutes progressing from easy to a few short openers, then a brief pause. Skip the 20-minute band sequence. If a stretch feels necessary, keep it under 60 seconds per muscle and follow it with pedalling. Do not copy a track-sprint warm-up that is itself a workout.

Uncertainty: Most stretch meta-analyses mix untrained and trained people and test jumps or isolated force, not a cycling time trial. The Hajoglou wattage gap is an access limit, not a claim that the two warm-ups were identical.

STR-C08 — Cycling injury prevention by strength, mobility, or activation is not established

Classification: Mostly coaching practice for the claim that these routines prevent cycling injuries or correct imbalances; Moderate evidence that strength training reduces injuries in other sports, with poor transfer, and that exercise programmes do not clearly reduce running injuries

Claim: Overuse knee and low-back pain are common in road cyclists, and crashes cause many of the recorded injuries. Neither fact is evidence that squats, planks, bands, or a mobility flow prevent those problems. No cyclist randomised trial retrieved here shows that heavy strength, core training, or an activation routine reduces cycling injury incidence. A FIFA 11+ warm-up study in cyclists is too flawed to carry that claim.

Support: Clarsen, Krosshaug, and Bahr (2010) interviewed 109 of 116 professional road cyclists. Of 94 overuse injuries, 45% were low back and 23% were knee; 58% had had low-back pain and 36% anterior knee pain in 12 months. Knee injuries caused more time loss. Fallon et al. (2025) synthesised competitive cycling epidemiology: upper-limb injuries, lacerations, abrasions, and fractures dominate several disciplines. Those are not squat-preventable events. Lauersen et al. (2014) found strength training reduced sports injuries in randomised trials (RR 0.315, 95% CI 0.207 to 0.480) and stretching did not (RR 0.963). Lauersen, Andersen, and Andersen (2018) updated the strength estimate (cluster-adjusted RR 0.338, 95% CI 0.238 to 0.480) across 7738 participants in non-cycling sports. Wu et al. (2024) found that exercise-based prevention programmes did not reduce running-injury risk or rate overall (1904 runners); a post hoc look at supervised programmes was more favourable. If the closest endurance sport does not show a clear effect, a cycling claim needs its own trial. Herman et al. (2012) supported some neuromuscular warm-ups for lower-limb injury in field and court sports. That is not a 5-minute glute circuit and not cycling. Kotler, Babu, and Robidoux (2016) recommend early correction of strength and flexibility imbalances, technique, and bike fit. That is clinical commentary. Bini, Hume, and Croft (2011) reviewed saddle height and knee risk and found disagreement about the method, not a prevention trial.

Contradiction: Panchal et al. (2025; doi:10.1038/s41598-025-91005-z) compared an 8-week FIFA 11+ warm-up (20 minutes, four days per week) with a conventional warm-up in male competitive cyclists and concluded that FIFA 11+ reduced injuries and improved performance. The full text describes a pre–post experimental design, later called quasi-experimental, with no non-intervention control. Methods say 79 enrolled and 10 dropped before inclusion (69 remaining); the abstract says 80. The a priori sample-size calculation asked for 104. Injury “incidence” was injuries divided by cyclists, not by exposure hours. The results narrative also refers to 3-kilometre running performance. The authors’ own limitations say long-term injury effects remain unclear. This paper does not establish cycling injury prevention. A 2025–2026 search did not retrieve a systematic review in which band glute activation improved cycling time-trial power or prevented cycling injury. Unverified percentage claims of that sort should be ignored.

Practical implication: Strength training is justified for performance and for general musculoskeletal capacity, not because a cyclist RCT has proved fewer crashes or fewer overuse injuries. Use bike fit, load management, and a clinician for persistent pain. There is no adequate evidence that activation, core work, or a mobility flow corrects a claimed imbalance or prevents cycling injury. Do not report that an imbalance has been corrected.

Uncertainty: Absence of cyclist prevention trials is not evidence that strength is useless for long-term tissue capacity. It is evidence that the resilience claim is ahead of the data. Bone health is a separate rationale reported by masters and professional cyclists; it is not tested in the performance trials above.

Heavy strength and what it does not change

Heavy strength here means lower-body loading at or above about 80% of 1RM, or the 4–10RM range used in the Rønnestad trials. It is not a band walk, a plank, or a light “activation” set.

Llanos-Lagos et al. searched PubMed, Web of Science, and Scopus to February 2025. Population: endurance cyclists aged at least 18, any sex or performance level. Comparator: cycling without heavy strength. Outcomes: VO₂max, power at VO₂max, maximal metabolic steady state (lactate steady state, second threshold, or critical power), efficiency or economy, anaerobic power and capacity, and performance efforts longer than 75 seconds. Seventeen studies, 262 participants, interventions of 5–25 weeks and 1–3 sessions per week. Full text was retrieved.

Outcome Pooled effect versus control Reading
Cycling performance (time trial and time to exhaustion combined) g 0.463 (0.109 to 0.817), p = 0.016, I² ≈ 0 Small-to-moderate benefit. Low certainty.
Cycling efficiency Small beneficial effect, magnitude about absolute g 0.353, p = 0.012, low certainty. Prose also prints point 0.353 with CI −0.619 to −0.088. Figure 3 shows a negative pool; caption: negative = beneficial. Source reporting inconsistency. Do not treat +0.353 with that negative CI as one signed estimate, and do not invent a replacement.
Anaerobic power (peak) g 0.560 (0.097 to 1.023), p = 0.024, I² = 29 Moderate benefit.
Anaerobic capacity (mean sprint power) g 0.235 (−0.230 to 0.700), p = 0.263 Not significant.
VO₂max g −0.041 (−0.345 to 0.263), p = 0.773 No effect.
Power at VO₂max g 0.164 (−0.176 to 0.505), p = 0.308 No effect.
Maximal metabolic steady state g 0.069 (−0.323 to 0.460), p = 0.697 No effect.

No moderator — sex, body mass, height, performance level, strength experience, or the intervention characteristics they tested — was significant. The practical section of the paper still calls heavy strength an effective strategy and points to a summary table of included programmes. It also says low certainty precludes a robust optimal prescription. Both sentences should be kept.

Sitko et al. (2024) is easy to misread on VO₂max. They estimated it from 5-minute power. The strength group’s larger “VO₂max” change is the 5-minute power result restated. It does not overturn the gas-exchange null in Llanos-Lagos et al.

Javaloyes et al. (first published 16 September 2025; journal year 2026) found that 1RM parallel squat was associated with maximal aerobic power in 69 national-team cyclists (r = 0.73) but not with time to exhaustion at that power. That is cross-sectional. Stronger riders are not, on that evidence, automatically better at holding maximal aerobic power.

Economy, sprint, time trial, and VO₂max

These qualities are not exclusive systems. A heavier squat does not replace mitochondrial or cardiac adaptation, and a VO₂ interval does not automatically raise half-squat 1RM. The useful model is interaction: more force per fibre, a slightly lower oxygen cost at a given power, and more peak power available for a jump or a short rise, sitting on top of whatever VO₂max and threshold the cycling has built.

Evidence capsules

Study Who, design, dose What changed What did not Access
Rønnestad, Hansen, Raastad 2010, Eur J Appl Physiol 20 well-trained cyclists; endurance plus strength n = 11 men versus endurance n = 9 (7 men, 2 women); 12 weeks; 4 lower-body exercises, 3 × 4–10RM, 2×/week. Not clearly randomised in the abstract. Thigh size, isometric force, Wingate peak power, power at 2 mmol·L⁻¹, and 40-minute mean power improved in the strength group. VO₂max rose similarly in both groups. Body mass did not rise with the larger thigh. Between-group superiority for the 40-minute test is not given as a p value; endurance-only tended to improve (P = 0.057). Abstract
Same group, 2011, Scand J Med Sci Sports Same sample and dose. 5-minute all-out after 185 minutes at 44% of maximal aerobic power. Companion paper, not an independent replication. 5-minute power 371 ± 9 to 400 ± 13 W in the strength group; no change in endurance-only. Lower oxygen cost, heart rate, lactate, and perceived exertion late in the long ride. Not a stochastic 3000 kJ race. Abstract
Rønnestad, Hansen, Raastad 2010 maintenance n = 6 versus 6. 12-week build at 2×/week, then 13 competition weeks at 1×/week. Size and 1RM preserved. From the start of the build to mid-competition, Wingate peak, power at 2 mmol·L⁻¹, maximal aerobic power, and 40-minute power rose; the last two improved more than endurance-only. Very small. One woman in the control arm. Abstract
Sunde et al. 2010 13 completers of 16 competitive road cyclists (10 men, 3 women); 4 × 4RM half-squat, 3×/week, 8 weeks, added to usual endurance. 1RM +14.2%, rate of force development +16.7%, economy at 70% VO₂max +4.8%, work efficiency +4.7%, time to exhaustion at old maximal aerobic power +17.2%. VO₂max and body weight unchanged. Control efficiency rose 1.4%, less than the intervention. Abstract
Aagaard et al. 2011 Young elite cyclists, n = 14 total; group sizes not in the retrieved abstract; 16 weeks concurrent strength. MVC and rate of force development +12–20% with strength. 45-minute capacity +8% with strength only. 5-minute performance +3–4% in both arms. IIA up, IIX down. VO₂max unchanged. Economy improved in the endurance arm until it matched strength. Capillarisation unchanged. Abstract
Rønnestad et al. 2015 Young elite; strength n = 9 versus endurance n = 7; 25 weeks heavy strength. Isometric squat, lower-body lean mass, Wingate peak, maximal aerobic power, power at 4 mmol·L⁻¹, 40-minute power, earlier peak torque. 40-minute gain correlated with earlier peak torque (r = 0.66). VO₂max and economy unchanged. Abstract
Rønnestad, Hansen, Nygaard 2017 Elite; strength n = 12 versus endurance n = 8; 10 weeks. Larger gains in isometric half-squat and 30-second Wingate mean power. Tendency for power at 4 mmol·L⁻¹ (P = 0.068). 40-minute all-out not significantly different (4 ± 6% versus −1 ± 6%, P = 0.13). Abstract
Vikmoen et al. 2016 19 well-trained women; randomised; 11 weeks; 2×/week heavy strength (n = 11) versus endurance (n = 8). 40-minute power, fractional utilisation, economy, leg-press strength, quadriceps size, IIX toward IIA. 40-minute change correlated with the fibre shift (r = −0.63) and size (r = 0.73). Men were not in this trial. Abstract
Vikmoen et al. 2017 19 well-trained female duathletes, VO₂max about 53–54 ml·kg⁻¹·min⁻¹; same 11-week dose; n = 11 versus 8. Related group and protocol; independence from the 2016 cohort is not established by the abstracts. Half-squat 1RM +45 ± 22% and leg lean mass +3.1 ± 4.0% more than endurance-only. 5-minute power after 3 hours of cycling +7.0 ± 4.5% in the strength arm; no change in controls. Running 5-minute power after 1.5 hours also rose, and that between-group contrast is explicit. Oxygen cost fell in the last 2 hours of the long ride. The abstract does not state a between-group p value for the cycling 5-minute test. Prolonged running economy did not change. Abstract; full text exists, not fully re-read
Sitko et al. 2024 36 trained road cyclists, mixed licence levels; 12 weeks preparatory phase; cycling volume matched (about 12 hours of prescribed zone-2 riding). Strength superior to core and to cycling-only across 5 seconds, 60 seconds, 5 minutes, and 20 minutes. 20-minute change about +0.22 W·kg⁻¹ versus +0.07 and +0.06. Core only helped 5-second power versus a control group that got slower. Body mass did not differ. Not blinded. Full text
Bishop et al. 1999 21 endurance-trained women, 18–42 years; squat 5 × 2–8RM to failure, 2×/week, 12 weeks; n = 14 versus 7. Squat 1RM +35.9% versus +3.7%. Authors concluded that the strength gain did not improve cycle endurance performance. Exact 1-hour power and lactate-threshold changes were stripped from the retrieved abstract. Abstract
Jackson et al. 2007 23 club cyclists; 10 weeks; high-load/low-rep, low-load/high-rep, or cycling only, same cycling plan. Strength rose in the lifting groups. No group difference on the lactate profile or economy. Outcome was not a time trial. Abstract

Rønnestad’s 2010 performance paper and 2011 prolonged-ride paper should be cited once as a programme, not as two discoveries. The prolonged-ride result is still the best cyclist evidence that heavy strength can improve a hard effort after a long steady ride. It is not evidence about durability after several stochastic hours at race pace. That question belongs with the durability topic.

Montalvo-Pérez et al. (2021) randomised 17 competitive women to 6 weeks of traditional or velocity-based lifting (squat, hip thrust, split squat, 2×/week) on top of usual endurance. Both improved strength. Velocity-based training was better for hip-thrust strength and power. Endurance outcomes did not differ between lifting styles. Six weeks without an endurance-only arm cannot carry the performance claim; it does show that the loading method is a second-order choice.

Louis et al. (2012) had 9 masters (51.5 ± 5.5 years) and 8 younger endurance athletes do 3 weeks of knee extensions, 10 × 10 at 70% of maximal repetition, 3×/week. There was no endurance-only control. Masters improved delta efficiency and gained more torque (+17.8% versus +5.9%). That is a short, moderate-load, single-exercise study. It suggests masters are not immune to a strength stimulus. It does not set the programme.

Hausswirth et al. (2010) assigned 14 triathletes to 5 weeks of added strength, 3×/week, or endurance only. The retrieved abstract supports a qualitative finding: combined training prevented the usual fall in freely chosen cadence over 2 hours. Exact physiological deltas were not recoverable.

Explosive work, plyometrics, and on-bike force

Three different ideas get collapsed in coaching language.

  1. Heavy strength with a fast concentric intent. This is still heavy strength. It is the Rønnestad model and the better-supported road-cycling option.
  2. Plyometrics. Paton and Hopkins mixed jumps with high-resistance cycling sprints. Gains were large and fast, in-season, and partly an efficiency and threshold story, but the sprints were cycling. Do not prescribe 20 single-leg jumps per side as a stand-alone economy tool on that paper.
  3. On-bike high-force pedalling. Barranco-Gil et al. (online 3 September 2024; Med Sci Sports Exerc 1 February 2025) randomised well-trained men to 10 weeks of full squats or all-out very-low-cadence efforts, both at 70% of mode-specific 1RM, 2×/week, with cycling volume and intensity matched, plus a control group (n = 12, 12, and 13). VO₂max did not change. Both lifting methods beat control for squat and pedalling strength (off-bike about +2.6% to +5.8%, on-bike about +4.5% to +7.3%, control about −5.8% to −3.9%) and for Wingate performance (+4.1% and +4.3% versus −4.9%). Quadriceps cross-sectional area rose about 2.5% and 2.2% versus −2.3%. Body composition did not differ. Pallarés et al. (online 5 February 2025) report the same group sizes, duration, and protocol, with extra outcomes: weekly cycling was about 10.7–10.9 hours, baseline VO₂max about 62–63 ml·kg⁻¹·min⁻¹, maximal aerobic power rose within the resistance arms, and time to exhaustion at respiratory-compensation-point power had effect sizes of at least 0.30 that were not significant. Efficiency had no significant group-by-time interaction. The on-bike arm uniquely improved pedalling-specific force. Stopping resistance training while continuing to ride was associated with loss of force and quadriceps size. The Biology of Sport title — cyclists do not need off-bike lifting — overstates a substitution result. The on-bike method was strength training: maximal intended force at about 70% of maximal dynamic force, not an easy spin. It has not been shown to match 4–10RM gym work for a 40-minute time trial. Count it as one programme, not two replications. A restored web search described the second paper as additional outcomes from the same intervention; the retrieved Pallarés text does not use that sentence, but the design match is too close to treat them as independent.

Vicari et al. (2026; first published 1 January 2026) compared 12 weeks of gym strength with on-bike specific strength, both twice weekly, in 19 young off-road cyclists. Gym training increased 30-second Wingate mean power (P = 0.002); the on-bike arm did not (P = 0.276). The sample is small and mixed by sex. This does not cancel Barranco-Gil. It warns that “on the bike” is not automatically equal to a heavy squat unless the force is actually high.

Kristoffersen et al. remain the clean specificity result: after a shared strength base, more heavy strength builds the squat, and short sprints build the sprint. Neither uniquely moved endurance capacity over the next 6 weeks.

Core and low-load activation

Sitko et al. are the direct test. Conventional strength improved the whole power curve. Core work did not, except a sprint comparison against a group that stopped sprinting. The core dose was not trivial: glute bridge and prone back extension for 10 slow repetitions, planks of 30 seconds, eight sets, twice weekly, for 12 weeks. If that does not move 20-minute power, a shorter daily activation sequence should not be assumed to.

Abt et al. show that a fatigued trunk can change knee kinematics. The inference “therefore train the core to prevent knee injury” skips the trial that would have to show fewer injuries. Pedal forces did not change in that experiment.

Activation EMG, including Distefano et al., answers “which exercise recruits gluteus medius?”. It does not answer “does this prevent patellofemoral pain in cyclists?” or “does this raise FTP?”. Crow et al. are an acute jump study in football. Transfer to a seated 20-minute effort is speculative.

No retrieved cyclist trial shows that a resistance-band warm-up corrects a claimed imbalance or prevents injury. “Imbalance” is not an operational outcome in these papers. Absence of that evidence is not proof that every band exercise is useless. Do not write a correction into a programme as if it had been measured.

Interference, spacing, soreness, progression

Separate three clocks.

Chronic interference. Endurance can shrink the strength, jump, and rate-of-force-development response (Rønnestad et al. 2012; Schumann et al. 2022 for explosive strength). It does not, in the cyclist meta-analysis, stop a performance benefit from appearing. Wilson et al. (2012) still matter historically: running interfered more than cycling, and more endurance frequency and duration predicted smaller strength effects. Held et al. (2026) reassure recreational concurrent trainees and warn that elite data are thin. A 7–12 hour cyclist is closer to the “trained but not 30-hour” trials than to untrained concurrent-training studies. Flag that transfer when quoting Schumann or Wilson.

Same-day order. Doing strength before endurance modestly protects dynamic strength (Eddens; Murlasits). Aerobic capacity does not care about the order in those meta-analyses. For an amateur whose goal is VO₂max and FTP, protect the cycling session. Put lifting after it, or on another day.

Acute damage. Devantier-Thomas et al. (2024) and the narrative review by Doma et al. (2019) agree that unaccustomed or muscle-damaging resistance work can raise the cost of the next rides and cut peak power for 24–48 hours. Introduce lifting in a week that does not also introduce a new VO₂ block. Expect the first 1–2 weeks to feel worse on the bike. That soreness is a programming problem, not a sign that strength “does not work for cyclists.”

Silva et al. (2014) are the opposite acute effect: heavy leg press minutes before a time trial can raise performance. Post-activation potentiation and next-day soreness are different phenomena. Do not use the 6.1% acute time-trial result to justify heavy legs the morning of a goal event until the athlete has tested it.

Progression. Successful cyclist protocols progress load, not exercise variety. When 4–8 or 4–10 repetitions are completed with the intended effort, add load. Sunde used 4 × 4RM three times per week; most Rønnestad blocks used twice weekly. Three times weekly is viable for a short 8-week economy block if cycling is not already maximal. It is not required. Llanos-Lagos et al. included 1–3 sessions per week and could not show frequency mattered. That is uncertainty, not a finding that once per week builds as fast as twice.

Maintenance. One heavy session per week preserved size and strength through 13 competition weeks in the small Rønnestad maintenance trial and was associated with further performance gains versus never lifting. Professional self-report (Vikestad et al. 2025) matches that seasonal pattern: two sessions off-season, one in-season, with travel, fatigue, and racing as the reasons they stop. Masters men report the same fatigue and time barriers, and older respondents more often cite bone health (Vikestad and Dalen, 2024). Those surveys do not prove efficacy.

On-bike substitute and cessation. In cyclists already riding about 11 hours, removing resistance training for 10 weeks coincided with losses of strength and muscle size, while either squats or matched high-force pedalling prevented that loss and raised Wingate power (Barranco-Gil abstract; Pallarés full text). If the gym is impossible, the substitute is hard low-cadence work with a strength purpose, scheduled like a strength session, not an extra hour of zone 2.

A 30–45 minute template and the 7–12 hour clock

This template is a synthesis of trial programmes, not a package that has itself been trialled, and not an optimal minute count. Two sessions per week in a build, and often one session per week for maintenance, are the supported implementation. Thirty to 45 minutes is a usable length taken from those sessions, not a proven optimum.

Build, 8–12 weeks, twice weekly

Block Content Time
Easy pedal or empty-bar sets Raise temperature. No band circuit. 5–8 min
Half-squat or bilateral leg press 3 × 4–8RM, about 2–3 min rest, controlled lower, intent to move the load quickly 12–15 min
Split squat or single-leg press 3 × 6–8RM per side 8–10 min
Hip hinge or single-leg hip extension, plus a calf raise 2–3 × 6–10 8–10 min
Optional 2–3 × 3–5 squat jumps or step-up jumps only after heavy work is tolerated and soreness is low 0–5 min

Total about 30–45 minutes including rests. Skip the jumps if they leave the next ride heavy. Do not add a fourth day of core work to “complete” the session.

Maintenance: often one session weekly once the build has landed, or two shorter sessions if the week is quiet and recovery allows. Keep at least one heavy lower-body exercise. If load cannot rise, hold the load. Do not turn maintenance into metabolic conditioning. One session is a supported way to hold the gain, not a rule that 30 minutes is best.

Where it sits in the week

  • Prefer the day after a hard ride, or a day with only easy cycling, not the day before the key VO₂ or threshold session during the first month.
  • Same-day pairing: quality ride first, lift at least several hours later, or lift in the morning only if the ride is easy. The 6-hour figure is a non-cyclist hedge (Robineau), not a cyclist law. A separate day is better when the calendar allows it.
  • Do not cut the only weekly intensity session to make room. If total hours are fixed, replace easy cycling minutes. If the quoted hours were cycling only, the strength time is an addition. Either way, record strength separately from the ride file so recovery cost is visible.

Time budget — strength minutes and recovery cost are explicit

Two readings of “7, 9, or 12 hours” are in use. The platform’s integrated weeks use total available time, including strength. If someone instead meant cycling-only hours, the same strength work adds time.

Reading 7 h 9 h 12 h
Total available time, including strength 390 min cycling + 30 min strength 480 min cycling + 60 min strength 660 min cycling + 60 min strength
If those hours were cycling only Add about 30–45 min for one maintenance session, or about 1–1.5 h for two build sessions Same addition Same addition
What not to do Do not drop the only hard ride to “fit” strength. Swap easy cycling if the ceiling is total time. Same Same

A two-session build does not fit the 30-minute slot in the 7-hour example without replacing more easy riding, or without treating 7 hours as cycling-only and adding the extra hour. That is a scheduling choice, not a finding that 30 minutes is optimal. Recovery cost is the soreness and the next-day peak-power dip, not only the clock. Count both.

For a 20–30 hour week the same session is a smaller fraction of load and a larger scheduling problem. Two sessions in preparation and often one in a race block remain the supported pattern. High endurance volume already blunts the strength response. Do not bury the session inside a 5-hour ride and call the whole file endurance.

An on-bike high-force substitute still counts as strength time, not as zone-2 volume.

What not to add: daily 15-minute activation, a separate core day, long static stretching before the strength session, and failure sets on every exercise in the first fortnight. The first fortnight should be technically heavy but not maximally sore.

Injury resilience and safety

Claim Status
Heavy strength improves trained cyclists’ time-trial or time-to-exhaustion performance Supported, moderate, low meta-analytic certainty
It raises VO₂max Not supported
It always raises economy or threshold Not supported; sometimes yes
It raises sprint peak power Supported more consistently than threshold
Core or band activation is required for those gains Not supported
Strength, core, stretch, or FIFA 11+ prevents cycling overuse or crash injury Not established in adequate cyclist trials
Strength reduces injury in some non-cycling sports Supported, poor transfer
Exercise programmes prevent running injury Overall null in Wu et al. 2024
Persistent pain means an imbalance that activation has corrected No adequate evidence for that correction or for cycling-injury prevention. Not a finding that every exercise is ineffective.

General safety, without turning this into a medical protocol:

  • Progress load. The injury-prevention literature outside cycling associates higher strength dose with lower injury risk (Lauersen 2018). That is not a licence to add volume until the knees ache.
  • Sharp joint pain, swelling, giving way, saddle sores that are worsening, hand numbness, or radicular pain: stop the offending stress and get a clinician. Bike fit, especially saddle height, is a more relevant first check than a new mobility flow (Bini et al. 2011), and even that literature does not agree on one formula.
  • Neck and hand symptoms in long rides are often position and duration problems. A plank does not fix a stem that is too long.
  • Strength training in the included cyclist trials did not systematically inflate body mass (Rønnestad 2010; Sunde 2010; Sitko 2024). Fear of “getting heavy” is not a reason to avoid the stimulus. Watch the scale for two weeks if that fear is personal; do not pre-emptively use tiny weights.
  • Pallarés et al. observed non-significant trends toward more lower-limb pain and stiffness in the squat arm (effect sizes 0.33 and 0.37). On-bike high-force work did not show that trend. If squats irritate a knee, the on-bike substitute is a rational swap, not a moral failure.
  • Crash injury is a handling, attention, equipment, and environment problem (Fallon et al. 2025). Do not sell strength training as crash prevention.

Long-term resilience — fewer overuse seasons across years — is plausible and unproven in cyclists. Say that plainly.

Warm-ups that change output

Separate the acute test from the chronic adaptation.

Question Best current answer
Does a cycling warm-up raise short performance versus sitting still? Yes, in small well-trained samples (Hajoglou; Burnley for moderate or heavy primes).
Is harder always better? No. Burnley’s sprint prime did not beat no prime. Tomaras’s long sprint warm-up was worse than a shorter one.
Is a hot bath equal to riding? For one 12-minute time trial in tier 2–3 cyclists, 45 minutes at 42 °C matched active warm-up and adding both did not help (Jaén-Carrillo 2026). Impractical, and not a reason to skip openers if you can ride.
Does foam rolling add to a cycling warm-up? Earp et al. (first published 22 July 2026) found no extra fall in biceps femoris stiffness versus cycling alone in 21 healthy active adults, and no performance test. Do not prescribe it for watts. Warneke et al. (2024, foam-rolling meta-analysis) found stretching and foam rolling were not superior to other warm-ups for acute flexibility.
Does a heavy lift immediately before a time trial help? It can (Silva 2014). It can also fatigue. Test it in training if used at all. It is not the warm-up.
Does an 8-week FIFA 11+ warm-up prevent cycling injury? Not established. Panchal et al. 2025 is quasi-experimental, internally inconsistent, and not exposure-adjusted.

A usable pre-race or pre-interval warm-up for this audience:

  1. 8–15 minutes easy to moderate cycling.
  2. Two or three openers of 10–30 seconds, not a full interval session, with easy pedalling between them.
  3. 2–5 minutes easy, then start. The gap should not be so long that muscle temperature falls; Faulkner and colleagues’ sprint-cycling work is about that problem, but the retrieved abstract here does not support a specific insulation protocol.
  4. Optional: one or two stretches held well under 60 seconds if a joint feels restricted, then more pedalling.

Skip: 50-minute warm-ups with repeated race-effort sprints; long static holds of the hip flexors and calves immediately before a maximal effort; a compulsory glute sequence.

Glutes, bands, stretch, and mobility

Glute activation and bands. No 2024–2026 systematic review retrieved here tests band glute warm-ups for cycling performance or cycling injury. EMG papers show that bands and side-lying exercises can recruit gluteal muscle. Recruitment is not performance and not prevention. A short band routine is a preference, not a requirement. It becomes false precision when it displaces sleep, food, or the actual warm-up.

Dynamic stretching. Opplert and Babault (2018) reviewed acute dynamic stretching and found range-of-motion gains and often small positive effects on force, power, sprint, and jump when the stretching is not excessive. That literature is mostly not cycling. Dynamic leg swings that mimic getting on the bike are harmless and may help range. They are not a separate training adaptation.

Static stretching. The dose result is stable from Kay and Blazevich (2012) through Behm et al. (2016), Chaabene et al. (2019), and Warneke and Lohmann (2024). Holds at or beyond about 60 seconds per muscle can reduce isolated maximal force. Shorter holds, especially after aerobic work and before sport-specific movement, usually do not matter for jumps and sprints. Cyclists are under-represented. The practical transfer is still: do not spend the warm-up in long static positions. Chronic stretching can increase range of motion; that is a different question from pre-ride performance and was not the focus of the cyclist trials.

Injury and stretching. Small et al. (2008) found the included randomised and controlled trials did not support static stretching in the warm-up as injury prevention. Lauersen et al. (2014) likewise found no stretching benefit and a strength-training benefit in non-cycling sports. Do not revive a long stretch routine for “injury resilience.”

Pre-ride mobility. A cyclist who cannot reach the bars without pain has a fit or a clinical problem. A cyclist who can ride the position does not need a daily mobility screen. Five minutes of easy pedalling moves the hips, knees, and ankles through the range that cycling uses. Extra mobility work is reasonable if a measured restriction changes fit or causes symptoms. It is not a performance pillar beside volume, intensity, and heavy strength.

Acute pain. A warm-up or a short stretch can reduce the feeling of stiffness. That analgesic or tolerance effect is not evidence that tissue was repaired or that an imbalance was corrected. If pain returns at the same point in the ride every time, change the position or the load and get it assessed.

7–12 hours versus 20–30 hours

The best-matching volume in the retrieved trials is the Barranco-Gil / Pallarés cohort at about 11 hours per week, VO₂max about 62 ml·kg⁻¹·min⁻¹, and the amateur end of Sitko’s mixed sample. Rønnestad’s “well-trained” cohorts and Vikmoen’s women are also closer to a serious amateur than to a 30-hour WorldTour week. Elite and young-elite trials (Aagaard; Rønnestad 2015 and 2017) still show neuromuscular and sometimes performance gains, but the 10-week elite 40-minute result was not statistically significant. More training age does not guarantee a larger time-trial dividend from the next 10 weeks of squats.

For 7–12 hours:

  • Two sessions in a build, and often one for maintenance, are a large enough fraction of a 7–12 hour total week to matter, and small enough to recover from if cycling has one or two hard days, not three. That is an implementation, not a 30-minute optimum.
  • The performance prize is more likely economy, short power, and a hard effort after steady work than a new VO₂max.
  • Activation time is the first thing to delete.
  • If 7–12 hours is total availability, strength replaces easy cycling: the integrated examples are 390+30, 480+60, and 660+60 minutes. If those hours were cycling only, a two-session build adds about 1–1.5 hours. Either reading must show the strength minutes and the recovery cost. Nine hours of cycling plus 80 minutes of lifting is a longer week than a 9-hour total budget.

For 20–30 hours:

  • The same gym session is a smaller fraction of load and a larger scheduling problem. Rønnestad et al. (2012) is the warning: high endurance volume blunts the strength adaptation. Pros still lift, mostly twice weekly off-season and once in-season (Vikestad et al. 2025), for performance, injury belief, and bone. The injury belief is ahead of cyclist trials.
  • Put the build in a preparation block. In a race block, one maintenance session beats a heroic twice-weekly plan that dies in travel.
  • Do not copy a 30-hour rider’s easy spinning and also copy a 30-hour rider’s omitted gym work, then add amateur-style intensity on top. The amateur’s scarce resource is high-quality cycling plus one recoverable strength dose, not more activation.

Neither group should treat strength, core, and mobility as three separate systems that each need their own microcycle. One heavy lower-body exposure, maintained, plus a cycling warm-up when the session is hard, covers what the evidence actually supports.

Open questions

  • Whether the Llanos-Lagos performance effect is mediated by efficiency, anaerobic power, or simply more high-quality training confidence. The review did not test mediation, and certainty is low.
  • The minimum dose for an amateur who has already lifted for years. Moderators were non-significant, which is not the same as “dose does not matter.”
  • A cyclist trial of 6 hours versus 24 hours between lifting and a key ride, with time-trial performance as the outcome.
  • Whether matched on-bike high-force work equals 4–10RM gym strength for 20–40 minute power. Current on-bike trials used about 70% and did not show a significant time-to-exhaustion gain.
  • A properly randomised, exposure-adjusted injury trial in cyclists. Until that exists, prevention language should stay off the programme sheet.
  • Sex-specific prescriptions. Women improved in Vikmoen et al.; Bishop et al. (1999) did not find an endurance-performance benefit from squat-only training to failure. Llanos-Lagos et al. found no sex moderator and included 60 women. Do not invent a different programme by sex from these data.
  • Durability after stochastic race loads. The 185-minute steady ride and the 3-hour duathlon ride are encouraging and incomplete.
  • Whether short static stretching inside a cycling warm-up changes 4–20 minute power at all. The stretch meta-analyses rarely test that.

What not to over-precision

Elaborate activation routines, the exact minute of a glute bridge, band colour, and the difference between a 4RM and a 6RM are small beside whether heavy loading happens twice a week for a couple of months and whether the hard cycling sessions stay recoverable. The transcript paraphrase — genuine strength rather than low-load activation — matches Sitko et al. and the absence of cyclist activation trials. It is still a paraphrase, not a cited quotation.

References

Full registry entries live in the shared bibliography. Links below are the originals used for this brief.

  • Llanos-Lagos C, Ramirez-Campillo R, Sáez de Villarreal E. Heavy strength training effects on physiological determinants of endurance cyclist performance: a systematic review with meta-analysis. Eur J Appl Physiol. 2026;126(1):193–222. Online 9 July 2025. doi:10.1007/s00421-025-05883-2 · PMC
  • Rønnestad BR, Hansen EA, Raastad T. Effect of heavy strength training on thigh muscle cross-sectional area, performance determinants, and performance in well-trained cyclists. Eur J Appl Physiol. 2010;108:965–975. doi:10.1007/s00421-009-1307-z
  • Rønnestad BR, Hansen EA, Raastad T. Strength training improves 5-min all-out performance following 185 min of cycling. Scand J Med Sci Sports. 2011;21:250–259. doi:10.1111/j.1600-0838.2009.01035.x
  • Rønnestad BR, Hansen EA, Raastad T. In-season strength maintenance training increases well-trained cyclists’ performance. Eur J Appl Physiol. 2010;110:1269–1282. doi:10.1007/s00421-010-1622-4
  • Rønnestad BR, Hansen EA, Raastad T. High volume of endurance training impairs adaptations to 12 weeks of strength training in well-trained endurance athletes. Eur J Appl Physiol. 2012;112:1457–1466. doi:10.1007/s00421-011-2112-z
  • Rønnestad BR, Hansen J, Hollan I, Ellefsen S. Strength training improves performance and pedaling characteristics in elite cyclists. Scand J Med Sci Sports. 2015;25:e89–e98. doi:10.1111/sms.12257
  • Rønnestad BR, Hansen J, Nygaard H. 10 weeks of heavy strength training improves performance-related measurements in elite cyclists. J Sports Sci. 2017;35:1435–1441. doi:10.1080/02640414.2016.1215499
  • Rønnestad BR, Mujika I. Optimizing strength training for running and cycling endurance performance: a review. Scand J Med Sci Sports. 2014;24:603–612. doi:10.1111/sms.12104
  • Sunde A, et al. Maximal strength training improves cycling economy in competitive cyclists. J Strength Cond Res. 2010;24:2157–2165. doi:10.1519/JSC.0b013e3181aeb16a
  • Aagaard P, et al. Effects of resistance training on endurance capacity and muscle fiber composition in young top-level cyclists. Scand J Med Sci Sports. 2011;21:e298–e307. doi:10.1111/j.1600-0838.2010.01283.x
  • Vikmoen O, et al. Strength training improves cycling performance, fractional utilization of VO₂max and cycling economy in female cyclists. Scand J Med Sci Sports. 2016;26:384–396. doi:10.1111/sms.12468
  • Vikmoen O, Rønnestad BR, Ellefsen S, Raastad T. Heavy strength training improves running and cycling performance following prolonged submaximal work in well-trained female athletes. Physiol Rep. 2017;5:e13149. doi:10.14814/phy2.13149
  • Sitko S, López-Laval I, Cirer-Sastre R. Influence of conventional resistance training compared to core exercises on road cycling power output. Cureus. 2024;16:e59371. doi:10.7759/cureus.59371
  • Barranco-Gil D, et al. Off- and on-bike resistance training in cyclists: a randomized controlled trial. Med Sci Sports Exerc. 2025;57:296–304. Epub 2024 Sep 3. doi:10.1249/MSS.0000000000003556
  • Pallarés JG, et al. Cyclists do not need to incorporate off-bike resistance training to increase strength, muscle-tendon structure, and pedaling performance. Biol Sport. 2025;42:185–195. doi:10.5114/biolsport.2025.146790 · PMC
  • Kristoffersen M, Sandbakk Ø, Rønnestad BR, Gundersen H. Comparison of short-sprint and heavy strength training on cycling performance. Front Physiol. 2019;10:1132. doi:10.3389/fphys.2019.01132
  • Paton CD, Hopkins WG. Combining explosive and high-resistance training improves performance in competitive cyclists. J Strength Cond Res. 2005;19:826–830. doi:10.1519/R-16334.1
  • Kordi M, et al. Cycling-specific isometric resistance training improves peak power output in elite sprint cyclists. Scand J Med Sci Sports. 2020. doi:10.1111/sms.13742
  • Vicari DSS, et al. Strength training in the gym versus specific strength training on the bike in young off-road cyclists. Br Med Bull. 2026. doi:10.1093/bmb/ldaf024
  • Beattie K, Carson BP, Lyons M, Kenny IC. The effect of maximal- and explosive-strength training on performance indicators in cyclists. Int J Sports Physiol Perform. 2017;12:470–480. doi:10.1123/ijspp.2016-0015
  • Bishop D, Jenkins DG, Mackinnon LT, McEniery M, Carey MF. The effects of strength training on endurance performance and muscle characteristics. Med Sci Sports Exerc. 1999;31:886–891. doi:10.1097/00005768-199906000-00018
  • Jackson NP, Hickey MS, Reiser RF. High resistance/low repetition vs. low resistance/high repetition training: effects on performance of trained cyclists. J Strength Cond Res. 2007;21:289–295. doi:10.1519/R-18465.1
  • Yamamoto LM, et al. The effects of resistance training on road cycling performance among highly trained cyclists: a systematic review. J Strength Cond Res. 2010;24:560–566. doi:10.1519/JSC.0b013e3181c86583
  • Berryman N, et al. Strength training for middle- and long-distance performance: a meta-analysis. Int J Sports Physiol Perform. 2018;13:57–63. doi:10.1123/ijspp.2017-0032
  • Montalvo-Pérez A, et al. Traditional versus velocity-based resistance training in competitive female cyclists. Front Physiol. 2021;12:586113. doi:10.3389/fphys.2021.586113
  • Louis J, Hausswirth C, Easthope C, Brisswalter J. Strength training improves cycling efficiency in master endurance athletes. Eur J Appl Physiol. 2012;112:631–640. doi:10.1007/s00421-011-2013-1
  • Hausswirth C, et al. Endurance and strength training effects on physiological and muscular parameters during prolonged cycling. J Electromyogr Kinesiol. 2010;20:330–339. doi:10.1016/j.jelekin.2009.04.008
  • Javaloyes A, et al. Strength as a predictor of aerobic power in competitive cyclists: national team insights. Int J Sports Med. 2026. First published 16 September 2025. doi:10.1055/a-2702-4879
  • Schumann M, et al. Compatibility of concurrent aerobic and strength training for skeletal muscle size and function: an updated systematic review and meta-analysis. Sports Med. 2022;52:601–612. doi:10.1007/s40279-021-01587-7
  • Wilson JM, et al. Concurrent training: a meta-analysis examining interference of aerobic and resistance exercises. J Strength Cond Res. 2012;26:2293–2307. doi:10.1519/JSC.0b013e31823a3e2d
  • Robineau J, et al. Specific training effects of concurrent aerobic and strength exercises depend on recovery duration. J Strength Cond Res. 2016;30:672–683. doi:10.1519/JSC.0000000000000798
  • Eddens L, van Someren K, Howatson G. The role of intra-session exercise sequence in the interference effect: a systematic review with meta-analysis. Sports Med. 2018;48:177–188. doi:10.1007/s40279-017-0784-1
  • Murlasits Z, Kneffel Z, Thalib L. The physiological effects of concurrent strength and endurance training sequence: a systematic review and meta-analysis. J Sports Sci. 2018;36:1212–1219. doi:10.1080/02640414.2017.1364405
  • Held S, et al. Maximizing adaptations in concurrent training: an umbrella review of meta-analyses. Sports Med. 2026;56:1489–1512. Epub 2026 Feb 28. doi:10.1007/s40279-026-02401-y
  • Doma K, Deakin GB, Schumann M, Bentley DJ. Training considerations for optimising endurance development: an alternate concurrent training perspective. Sports Med. 2019;49:669–682. doi:10.1007/s40279-019-01072-2
  • Devantier-Thomas B, Deakin GB, Crowther F, Schumann M, Doma K. The impact of exercise-induced muscle damage on various cycling performance metrics: a systematic review and meta-analysis. J Strength Cond Res. 2024;38:1509–1525. doi:10.1519/JSC.0000000000004629
  • Silva RAS, et al. Acute prior heavy strength exercise bouts improve the 20-km cycling time trial performance. J Strength Cond Res. 2014;28:2513–2520. doi:10.1519/JSC.0000000000000442
  • Blazevich AJ, Babault N. Post-activation potentiation versus post-activation performance enhancement in humans. Front Physiol. 2019;10:1359. doi:10.3389/fphys.2019.01359
  • Vikestad V, Lyngstad IK, Dalen T. Strength training among professional UCI road cyclists: practices, challenges, and rationales. PLoS One. 2025;20:e0328195. doi:10.1371/journal.pone.0328195
  • Vikestad V, Dalen T. Strength training among male master cyclists—practices, challenges, and rationales. J Funct Morphol Kinesiol. 2024;9:232. doi:10.3390/jfmk9040232
  • Lauersen JB, Bertelsen DM, Andersen LB. The effectiveness of exercise interventions to prevent sports injuries: a systematic review and meta-analysis of randomised controlled trials. Br J Sports Med. 2014;48:871–877. doi:10.1136/bjsports-2013-092538
  • Lauersen JB, Andersen TE, Andersen LB. Strength training as superior, dose-dependent and safe prevention of acute and overuse sports injuries. Br J Sports Med. 2018;52:1557–1563. doi:10.1136/bjsports-2018-099078
  • Wu H, et al. Do exercise-based prevention programs reduce injury in endurance runners? A systematic review and meta-analysis. Sports Med. 2024. doi:10.1007/s40279-024-01993-7
  • Clarsen B, Krosshaug T, Bahr R. Overuse injuries in professional road cyclists. Am J Sports Med. 2010;38:2494–2501. doi:10.1177/0363546510376816
  • Fallon T, Palmer D, Bigard X, Heron N. Epidemiology of injury and illness across all the competitive cycling disciplines: a systematic review and meta-analysis. BMJ Open Sport Exerc Med. 2025. doi:10.1136/bmjsem-2024-002364
  • Kotler DH, Babu AN, Robidoux G. Prevention, evaluation, and rehabilitation of cycling-related injury. Curr Sports Med Rep. 2016. doi:10.1249/JSR.0000000000000262
  • Bini R, Hume PA, Croft JL. Effects of bicycle saddle height on knee injury risk and cycling performance. Sports Med. 2011. doi:10.2165/11588740-000000000-00000
  • Abt JP, et al. Relationship between cycling mechanics and core stability. J Strength Cond Res. 2007;21:1300–1304. doi:10.1519/R-21846.1
  • Distefano LJ, Blackburn JT, Marshall SW, Padua DA. Gluteal muscle activation during common therapeutic exercises. J Orthop Sports Phys Ther. 2009. doi:10.2519/jospt.2009.2796
  • Crow JF, Buttifant D, Kearny SG, Hrysomallis C. Low load exercises targeting the gluteal muscle group acutely enhance explosive power output in elite athletes. J Strength Cond Res. 2012. doi:10.1519/JSC.0b013e318220dfab
  • Panchal R, et al. Comparing the effectiveness of the FIFA 11+ warm-up and conventional warm-up in enhancing cyclist performance and mitigating injury risk. Sci Rep. 2025;15:9430. doi:10.1038/s41598-025-91005-z
  • Herman K, Barton C, Malliaras P, Morrissey D. The effectiveness of neuromuscular warm-up strategies that require no additional equipment for preventing lower limb injuries during sports participation: a systematic review. BMC Med. 2012;10:75. doi:10.1186/1741-7015-10-75
  • Burnley M, Doust JH, Jones AM. Effects of prior warm-up regime on severe-intensity cycling performance. Med Sci Sports Exerc. 2005;37:838–845. doi:10.1249/01.MSS.0000162617.18250.77
  • Hajoglou A, et al. Effect of warm-up on cycle time trial performance. Med Sci Sports Exerc. 2005;37:1608–1614. doi:10.1249/01.MSS.0000177589.02381.0A
  • Tomaras EK, MacIntosh BR. Less is more: standard warm-up causes fatigue and less warm-up permits greater cycling power output. J Appl Physiol. 2011:228–235. doi:10.1152/japplphysiol.00253.2011
  • Jaén-Carrillo D, et al. Passive warm-up via hot-water immersion provides similar cycling performance benefits to active warm-up during a 12-minute cycling time trial. Int J Sports Physiol Perform. 2026. First published 22 September 2026. doi:10.1123/ijspp.2026-0278
  • Earp JE, et al. The acute effects of a cycling warm-up with and without foam rolling on biceps femoris stiffness and tissue homogeneity. J Strength Cond Res. 2026;40:1238–1244. doi:10.1519/JSC.0000000000005513
  • Kay AD, Blazevich AJ. Effect of acute static stretch on maximal muscle performance: a systematic review. Med Sci Sports Exerc. 2012;44:154–164. doi:10.1249/MSS.0b013e318225cb27
  • Behm DG, Blazevich AJ, Kay AD, McHugh M. Acute effects of muscle stretching on physical performance, range of motion, and injury incidence in healthy active individuals: a systematic review. Appl Physiol Nutr Metab. 2016;41:1–11. doi:10.1139/apnm-2015-0235
  • Warneke K, Lohmann LH. Revisiting the stretch-induced force deficit: a systematic review with multilevel meta-analysis of acute effects. J Sport Health Sci. 2024;13:805–819. doi:10.1016/j.jshs.2024.05.002
  • Chaabene H, Behm DG, Negra Y, Granacher U. Acute effects of static stretching on muscle strength and power: an attempt to clarify previous caveats. Front Physiol. 2019;10:1468. doi:10.3389/fphys.2019.01468
  • Small K, Mc Naughton L, Matthews M. A systematic review into the efficacy of static stretching as part of a warm-up for the prevention of exercise-related injury. Res Sports Med. 2008. doi:10.1080/15438620802310784
  • Opplert J, Babault N. Acute effects of dynamic stretching on muscle flexibility and performance: an analysis of the current literature. Sports Med. 2018. doi:10.1007/s40279-017-0797-9
  • Warneke K, et al. Foam rolling and stretching do not provide superior acute flexibility and stiffness improvements compared to any other warm-up intervention: a systematic review with meta-analysis. J Sport Health Sci. 2024. doi:10.1016/j.jshs.2024.01.006