Research · Racquet Sports & Swimming

Same Motion, Every Rep: What the Evidence Says About Overuse Injury in Youth Tennis, Badminton, and Swimming

By Muhammad Dzulhisham · MSc Sports Coaching, NSCA-CSCS

Two overuse-injury stories already have real evidence behind them on this site: what happens when a young athlete stays in one team sport year-round, and what happens to a growing elbow when a young athlete throws — overwhelmingly in baseball — at high volume. Tennis, badminton, and competitive swimming don't fit cleanly into either story, and treating them as smaller versions of the same problem misses what the literature actually shows about how they injure young bodies.

A team-sport specialist's joints absorb a mix of movements even inside one sport — a footballer's knee cuts, jumps, kicks, and sprints in different ratios every session. A throwing specialist's risk, by contrast, concentrates in a single, explosive, individually countable action — a pitch — which is exactly why baseball built an entire guideline system, Pitch Smart, around counting it. Racquet sports and swimming sit in a third category the evidence treats differently again: each asks one joint, almost always the shoulder, to repeat close to the same motion many hundreds or thousands of times inside a single session, with no equivalent of a pitch count ever built around the exposure — and in tennis and badminton specifically, a second joint, the wrist, absorbing much of the same repetitive load at the far end of the same swing.

Tennis: where the shoulder, elbow, and wrist injuries actually cluster

A 2019 systematic review in Translational Sports Medicine, covering overuse injury at the elbow and shoulder across youth overhead sports, found incidence rates highest in exactly three sports: tennis, baseball, and softball¹ — placing junior tennis in the same overuse-risk bracket as the throwing sports already covered elsewhere on this site, despite the very different motion involved.

A more recent, tennis-specific systematic review, published in Sports in 2025, puts real numbers against that ranking.² Competitive juniors sustained 2.11 to 3.50 injuries per 1000 player-hours, with 46–54% of juniors reporting at least one injury per season. Lower-limb trauma was the single largest category (48–54% of injuries), but a national squad cohort within the same review recorded shoulder problems accounting for up to 34% of all time-loss episodes, and one cohort reported 43% of injuries as severe, meaning more than 28 days out of play.

The review's risk-factor findings are specific enough to act on. An acute-to-chronic workload ratio above 1.3 — a sharp week-over-week jump in training load relative to an athlete's recent average — raised injury risk by a factor of 1.6 in the following week, and week-to-week training increases of 30% or more correlated specifically with lumbar and shoulder injuries. On the shoulder itself, a loss of glenohumeral internal rotation of 15 degrees or more — a rotational mobility deficit commonly abbreviated GIRD, and one of the best-documented risk markers in overhead athletes generally — doubled the risk of a symptomatic shoulder, alongside scapular dyskinesis and poor core stability as factors distinguishing injured players from uninjured peers.

The same review is direct about what actually reduced injuries when it was tested. A six-week core-stability programme cut overuse injury incidence from 0.41 to 0.30 per player-season — roughly a 27% reduction — and personalising racquet grip size to the individual player's hand cut the prevalence of lateral epicondylalgia (tennis elbow) from 14% to 6%, more than halving it with a single equipment adjustment.

A separate 2019 systematic review in the Journal of Sports Sciences, screening 4,067 articles down to five that met its inclusion criteria for risk factors specifically, adds a genuinely counterintuitive finding.³ It found moderate evidence that previous injury, regardless of body location, predicted future injury — unsurprising — but also moderate evidence that fewer years of tennis experience, not more, predicted upper-extremity injury specifically. That cuts against the instinct that cumulative career volume is always the dominant risk driver; technique and conditioning immaturity in newer players appears to carry real weight of its own, at least for the shoulder and elbow.

The wrist: a second joint doing the same repetitive work

Tennis loads the wrist through the same repetitive mechanism as the shoulder and elbow, and a 2016 case report in the Journal of Medical Case Reports documents exactly how.⁴ An elite junior player developed a scaphoid stress fracture — a bone injury more commonly associated with a single fall onto an outstretched hand — from nothing more dramatic than repeated practice of the attacking backhand high volley, a stroke that demands excessive dorsal flexion of the wrist at contact, applied often enough that the bone itself gave out before any single traumatic moment occurred. The player recovered fully with conservative treatment and returned to competition within five months, but the mechanism the case illustrates generalises well beyond one stroke: modern tennis technique asks the wrist for a wide range of motion across serves, backhands, and volleys alike, and that range gets used thousands of times a week in a training junior's schedule.

Badminton: the same overhead motion, starting younger

Badminton's overhead smash shares its basic mechanics with a tennis serve or a baseball pitch — a rapid arm-cocking phase followed by an explosive, decelerating strike — but a competitive match calls for it, and its lower-intensity overhead clears, far more times per session than either of those other sports asks of a single arm. The epidemiology tracks that difference in exposure.

A 2025 study in Scientific Reports, surveying 711 competitive players aged 7 to 22, found 60.3% had sustained at least one badminton-related injury.⁵ Injury rate peaked at 3.24 per 1000 training-hours in males aged 15–16 and 3.52 per 1000 training-hours in females aged 17–18. Knee and ankle injuries were the most common sites overall, but the shoulder still accounted for 6.5% of all injuries — 7.6% in males specifically, the fourth most common site in that group.

What stands out is how early that pattern starts. A separate 2021 study in the same journal surveyed 611 elementary school-aged players, ages 7 to 12, competing in Japan's national schoolchildren tournament.⁶ Among the 510 players with complete data, 8.0% reported shoulder pain — and the dose-response relationship was already measurable at that age: players training more than 2.5 hours a day were 2.64 times more likely to report shoulder pain than those training 2.5 hours or less (95% CI 1.03–6.78, p=0.043). Shoulder pain, lower-back pain, and knee pain also clustered together statistically, each strongly predicting the others — a pattern consistent with a single overloaded young athlete accumulating pain at more than one site rather than three unrelated, coincidental complaints.

A 2022 prospective cohort study in the International Journal of Environmental Research and Public Health, following 38 under-17 players from the Spanish Championship for 12 months, found a nearly identical shoulder-pain rate to the adult and adolescent literature above: 47.4% reported at least one episode.⁷ The single strongest predictor the study identified was, again, a loss of internal rotation range of motion on the dominant shoulder — the same GIRD-type deficit flagged in the tennis literature. A cutoff of 55 degrees or less predicted shoulder pain one year later with 75% sensitivity and 83.3% specificity, meaning a simple goniometer measurement, taken once, correctly flagged the great majority of players who would go on to develop shoulder pain over the following year.

A pitch has a name, a count, and an entire guideline system built around limiting it. A tennis serve, a badminton smash, and a competitive freestyle stroke are the same category of repeated, high-velocity shoulder motion — and none of them have ever had anything like that built around them.

Swimming: swimmer's shoulder, and a genuine complication in the evidence

Swimming removes impact and ground-reaction force almost entirely, which is part of why it's often recommended as the "safe" sport for a young athlete recovering from something else. It replaces that risk with a different one: an even more purely repetitive shoulder motion than either racquet sport, performed for far more total repetitions per session.

A 2024 multi-site survey in the International Journal of Sports Physical Therapy, covering 671 competitive swimmers aged 9 to 17 across six US states, found 49% had shoulder symptoms that affected their swimming.⁸ The pattern worsened with age within that same youth range: current shoulder pain was reported by 21.3% of 13-to-14-year-olds against 28.1% of 15-to-17-year-olds, and a history of pain that had interfered with swimming climbed from 47.4% to 55.5% across the same two age bands. Greater competitive experience correlated significantly with both more pain and more disability (p<0.001) — consistent with the shoulder accumulating measurable wear across a swimming career rather than the injury appearing at random.

A 2020 systematic review in the Journal of Athletic Training quantifies exactly how much training volume climbs across that same age range, and what happens to shoulder pain alongside it.⁹ Swimmers under 15 trained 6.64 to 12.23 hours a week; by ages 15 to 17, that rose to 12.31 to 17.27 hours a week. Adolescents in that 15-to-17 bracket showed the highest shoulder-pain prevalence of any age group studied, at 91.3%, against a range of 19.4% to 70.3% in every other age group measured. Within that adolescent group specifically, the review found moderate-quality evidence that swimmers training beyond roughly 15 hours or 35 kilometres a week faced measurably higher tendinopathy risk — and, in a detail that reads almost as a direct biological confirmation of the mechanism, every swimmer in the pooled data who showed ultrasound evidence of shoulder-tendon thickening also reported shoulder pain.

Here the evidence gets genuinely more complicated, and it's worth reporting honestly rather than smoothing over. A 2023 systematic review in the Scandinavian Journal of Medicine & Science in Sports, using a best-evidence synthesis across 22 studies and roughly 80 separate variables tested against shoulder pain in competitive swimmers, found no strong evidence either supporting or refuting the great majority of those variables.¹⁰ More strikingly, several of the factors that show up as clear risk markers in the tennis and badminton literature above — reduced shoulder range of motion, training frequency, height and weight, sex, and age — actually carried moderate evidence against an association with shoulder pain in swimmers specifically. The two factors that did carry moderate supporting evidence were a swimmer's competitive level and their shoulder muscle recruitment profile, particularly reduced activity in the serratus anterior, a key scapular-stabilising muscle. The review's own conclusion points toward shoulder strength-endurance, not passive range of motion or raw training hours, as the most clinically relevant modifiable factor in this specific sport.

That's a real, evidence-based distinction from the racquet-sport pattern above, not a contradiction to explain away. Swimming's stroke is submaximal and continuous rather than the single violent, decelerating action of a serve, smash, or pitch — a difference in mechanism significant enough that a screening and prevention framework built around GIRD and total-volume ceilings, well-supported in tennis and badminton, does not transfer cleanly onto swimmers on the current evidence. What transfers instead is a narrower, muscle-endurance-specific one.

What this means for how these athletes should actually train

Three sports, three slightly different evidence-based answers, but a shared starting point: whichever sport is involved, the joint doing the repeating needs to be measured, not assumed safe because the sport carries no contact and no obvious impact.

• Track internal rotation, not just pain, in tennis and badminton. A goniometer measurement of dominant-shoulder internal rotation is cheap, fast, and — per the badminton cohort above — predicted a year's worth of shoulder pain with better than 80% probability at a single cutoff.
• Treat sudden jumps in load as the trigger, not total volume alone. The tennis data's ACWR finding and the week-to-week increase threshold both point the same direction: a spike relative to what the shoulder has adapted to is the specific danger, echoing the same training-load caution documented elsewhere on this site for team sports, even though the underlying mechanism here is a single repeated joint action rather than a whole-body training programme.
• Build scapular and core strength as a standing part of the programme, not a rehab afterthought. A core-stability block cut tennis overuse injuries by roughly a quarter; scapular muscle recruitment, specifically serratus anterior function, is the one modifiable factor the swimming evidence actually supports.
• Don't assume swimming's prevention playbook matches tennis and badminton's. The mobility-screening and volume-ceiling approach that has real support in the racquet sports has not held up the same way in the swimming-specific literature — a genuine, evidence-based reason to build a swimmer's programme around strength-endurance rather than importing a framework built for a different motion.
• In tennis, check the equipment before assuming the problem is training. Grip-size personalisation more than halved lateral epicondylalgia prevalence in the cited review — a fix that costs nothing in training time.

None of this argues these sports are more dangerous than the team sports and throwing sports already covered on this site — a 49% shoulder-symptom rate in teenage swimmers and a 47–60% injury rate across youth tennis and badminton cohorts sit in a broadly similar range to the specialization data reported elsewhere. What the evidence does argue is that "no contact, no impact" is not the same claim as "no overuse risk," and that the specific joint, the specific motion, and — in swimming's case — the specific mechanism each demand their own evidence rather than a generic overuse-prevention template borrowed from a different sport entirely.