← All news and papers

How much is too much? What the evidence says about training a young athlete

14 November 2026 · Article · 18 minute read ·

In shortMost of the training limits parents are given for young athletes are conventions, not findings. What the evidence supports, and the five signs that matter.

This article is about what the research does and does not show. It is not medical advice and it is not a substitute for seeing a GP. Any persistent pain, or any concern about a child’s health, growth or eating, should go to a GP.

A parent who goes looking for guidance on how much training is enough for an 11 year old will find numbers quickly. No more hours a week than the child’s age in years. At least one or two days off a week. Two or three months a year away from the sport. No more than eight months a year in a single sport. The numbers are specific, they are repeated by governing bodies and clubs, and they sound as though someone measured something.

Mostly, nobody did. Three of those four figures have no derivation anywhere in the published literature, one of them began life as a line in a press release, and the one rule that has been tested properly stopped working when its own authors ran the full analysis. That is not an argument for ignoring volume, which turns out to matter a great deal. It is an argument for knowing which of the things you have been told is load-bearing, because a parent who worries about the wrong number will miss the signals that do mean something.

How much training is too much for a young athlete?

The honest answer is that total weekly hours of organised sport is the measure with evidence behind it, and no threshold within it has been established. The closest thing to a proper test followed young athletes recruited through two American hospital systems, and found that each additional five hours a week of organised sport was associated with about a 31% increase in the odds of injury, at 1.31 with an interval from 1.19 to 1.44, and the same 31% increase for overuse injury specifically, at 1.31 from 1.07 to 1.59 (Jayanthi et al., 2020). It is a clinical case-control design, just under half of those enrolled completed the first follow-up, and the sample was 53% female with a mean age of 14. A separate 52-week prospective study in Swedish youth athletics points the same way on training hours (Ek, Kowalski and Jacobsson, 2022). That is a gradient rather than a cliff edge. There is no point at which a child crosses from safe to unsafe, which is inconvenient for anyone wanting a rule and closer to how growing bodies behave.

The practical consequence for a parent is to add the hours up across everything. School PE, school teams, the club, the second sport, the Saturday fixture, and the private coaching all count. Most parents who are surprised by a total have been counting only the club.

Where does the “hours equals age” rule come from?

From a conference presentation in April 2013 and the press release that accompanied it, rather than from a paper. The release reported that athletes playing more hours a week than their age in one sport were 70% more likely to suffer serious overuse injuries than other injuries, and recommended that young athletes should not spend more hours a week in organised sport than their age.

It was then tested. A case-control study of 1,190 athletes aged 7 to 18 found that exceeding the threshold roughly doubled the odds of serious overuse injury, at 2.07 with an interval from 1.40 to 3.05 (Jayanthi et al., 2015). An independent test in 2,011 athletes aged 12 to 18 found a smaller association with any injury in the previous year, 1.34 from 1.12 to 1.61 (Post et al., 2017), and opened by stating that limited evidence existed to support or refute the recommendations in the first place. That study applied the threshold to hours in the child’s primary sport, where the original rule and the article you are reading concern total organised sport hours, so the two are not measuring quite the same thing. Both studies also asked athletes to recall a year of injuries, which tends to inflate associations.

The closest thing to a decisive test came from the same group that produced the rule. In their longitudinal follow-up, exceeding the age threshold was associated with overuse injury on the simple analysis, at 1.68 from 1.25 to 2.25 (Table 4). Then they added total weekly hours to the model. The threshold fell to 1.18, with an interval from 0.77 to 1.80 and a p value of 0.44, while total hours survived (Jayanthi et al., 2020, Table 5). In plain terms, the rule was picking up the fact that children who exceed it are doing more sport.

In fairness to the authors, they did not read it that way. Their discussion holds that exceeding the volume recommendations may continue to be a risk, and they go on recommending age-appropriate volume limits. A reader can look at Table 5 and reach a different conclusion, which is what I do here. Either way the rule is a useful way of talking about volume with a 12 year old rather than a boundary, and a child one hour under it is not safer than a child one hour over.

Do the rest day and months off rules hold up?

They are committee opinion, and the committee said so. The one or two rest days a week and the two to three months a year away from a sport both come from a 2007 American Academy of Pediatrics clinical report, which attaches no citation to either figure and states openly that there is a very small body of scientific evidence on these questions and that some of its recommendations are therefore based on committee opinion and expertise. The 2024 update repeats them, and manages to contradict itself in the process, with the body recommending at least one rest day and the clinician guidance recommending one to two.

No study has compared one rest day with two, or two months off with three, against any outcome at all. The eight month rule has a different and stranger history. It was never a threshold. It began as one of three criteria used to classify an athlete as specialised, alongside naming a main sport and having quit other sports, and it has never been compared with seven months or nine. Exceeding it is associated with overuse injury, at 1.66 to 1.68 for upper and lower limb in one cross-sectional study (Post et al., 2017) and 1.58 from 1.23 to 2.04 in the longitudinal one. Consistency requires me to add that the second of those is univariable, and nobody has tested whether it survives adjustment for total weekly hours. The eight-month rule has the same weakness I have just used against the hours rule, and it should be read the same way.

UK parents should know how these reach them. England Athletics’ Youth Endurance guidance of March 2022 passes both the eight month rule and the hours-equals-age rule to coaches, attributed to a 2012 paper that did not generate either of them. The same document carries a weekly running volume table by age, and is candid that its charts should be used as guidelines and adapted as necessary rather than followed as a prescription. A coach quoting those numbers at you is quoting a sensible practitioner document, not a measured safety limit.

Is early specialisation actually harmful?

Probably, modestly, and for reasons nobody has established. Two meta-analyses found that highly specialised young athletes carry a higher injury risk than those who sample several sports, with pooled risk ratios of 1.81 from 1.26 to 2.60 (Bell et al., 2018) and 1.37 from 1.19 to 1.57 (Carder et al., 2020). Those rest on four and six primary studies respectively, drawn from a largely overlapping pool, so their agreement reflects shared data rather than independent replication. A narrative review from inside the field finds risks associated with a highly specialised approach while noting that the mechanisms driving them are largely unknown, and the field cannot agree how to define specialisation in the first place (Mosher et al., 2022).

The stronger finding concerns performance, and it is the one most likely to change a parent’s mind. A meta-analysis of 51 study reports covering 6,096 athletes, including 772 of the world’s top performers, found that adult world-class athletes had done more multi-sport practice as children and adolescents, and had started their main sport later, than their less successful national-class peers (Güllich, Macnamara and Hambrick, 2022).

The same meta-analysis found the opposite pattern among junior athletes, where the better performers had started earlier and done more practice in their main sport. Both halves matter to a parent. Early specialisation does tend to produce the better 14 year old, and it is at 24 that the pattern reverses. That is an association read backwards from athletes who all reached national level, rather than a prediction about any individual child, but it does mean specialising early is not the price of becoming excellent.

Should I worry about the growth spurt?

Less than you have been told, and differently. The idea that the year around peak height velocity is an injury window rests substantially on two papers that share a single sample of 26 footballers. Set against that, the 1,190-athlete study above measured height directly and found growth rates identical in injured and uninjured athletes, 4.8 cm a year in both groups, concluding flatly that growth rate was not related to injury risk (Jayanthi et al., 2015). There is, on this evidence, no good reason to reduce a child’s training on the basis of growth alone. That study did not identify which athletes were at peak height velocity and did not test a change in training, so it cannot tell you what happens if you do adjust. What it does not license is ignoring pain at a growth area, which peaks at exactly this age, and decisions about an individual child belong with the coach and, where there is pain, a GP.

The specific growth-related conditions are a different matter, and here the folk wisdom is wrong in the other direction. Osgood-Schlatter disease, pain at the bony bump below the kneecap, occurs at about 3.8 cases per 1,000 person-years in children aged 8 to 18 in general practice, peaking at age 12 in boys (van Leeuwen et al., 2022). Children are routinely told they will grow out of it. A small cohort study enrolled 51 adolescents and found that, of the 46 still available two years on, 37% were still reporting knee pain, with a median duration since onset of 42 months in that subgroup (Holden et al., 2021). Most had recovered. Enough had not that the reassurance is too confident. Sever’s disease, the heel equivalent, appears at around 3.7 per 1,000 registered patients aged 6 to 17 (Wiegerinck et al., 2014). Pain at the front of the knee or the back of the heel that has lasted weeks warrants a GP appointment rather than patience. Naming the condition is the GP’s job rather than a parent’s.

As for the belief that children get clumsy and slower through the growth spurt, there is no validated measure of it and no strong longitudinal evidence. A review that screened 2,304 studies found only six addressing whether adolescence involves regression at all, and concluded cautiously that at least some children do show delays or regressions in some aspects of sensorimotor function (Quatman-Yates et al., 2012). Coaches are probably seeing something real in some children. The literature cannot yet tell you which ones.

One athletics finding bears directly on this, though it concerns elite athletes rather than club ones. Among 5,766 sprinters and 5,863 jumpers in the world all-time rankings, birth-quarter bias was present at under-18 level and absent among seniors, and the relatively younger athletes in an age group transitioned to senior level at a higher rate, with an odds ratio of about 1.64 (Brustio, Stival and Boccia, 2023). Being born late in the selection year looks like a disadvantage for junior selection and no disadvantage at all for senior success. Note that this is about birth date rather than physical maturity, which the study did not measure.

What about training load apps and workload ratios?

Treat any product promising to predict your child’s injury risk from a workload ratio with real scepticism. The acute-to-chronic workload ratio, which compares recent training with the preceding few weeks, has been tested to destruction in the methodological literature. One study rebuilt the calculation using randomly generated numbers in place of the real chronic workload and got much the same result, and neither version outperformed a model containing no information at all (Impellizzeri et al., 2021). The study usually cited as its youth evidence had a mean participant age of 17.3 years, and it reported workload ratios apparently predicting contact injuries, which a ratio cannot plausibly cause (Bowen et al., 2017). No blood marker, hormonal measure or heart-rate-variability figure has been shown to flag overtraining accurately in a young athlete either, and the international consensus statement on overtraining concludes that no single diagnostic marker exists (Meeusen et al., 2013).

What are the real warning signs?

Five things have published evidence behind them. They are unglamorous, and none of them requires an app.

There is a sixth, which belongs with a doctor rather than a coach. Low energy availability, now usually discussed as Relative Energy Deficiency in Sport following the International Olympic Committee’s 2023 consensus statement, presents with vague symptoms, is easily missed in boys, and can occur in a child of entirely ordinary build (Todd, Elliott and Keay, 2022). If a young athlete is training hard and something seems wrong, the route is a GP appointment, and from there a referral where it is needed. This is not territory for an app, a chart or a parent’s own numbers, and nothing in this article should be used in place of that appointment.

What does UK guidance actually say?

It sets a floor and no ceiling. The UK Chief Medical Officers’ guidelines ask children and young people aged 5 to 18 to average at least 60 minutes of moderate to vigorous activity a day across the week, which is a public health minimum designed for a largely inactive population. A parent who follows it faithfully still has no reference point for the question this article is asking, because on training volume for an 11 to 16 year old in community sport there is no UK dataset and no UK limit. The Long-Term Athlete Development model, which sits behind a good deal of what governing bodies tell coaches, has itself been criticised from inside paediatric exercise science for a thin evidence base, particularly its windows of opportunity (Ford et al., 2011), and the International Olympic Committee’s youth development statement is careful to call itself evidence-informed and to offer guiding principles rather than thresholds (Bergeron et al., 2015). Where a model hands you a number, ask where the number came from.

How does Athleet.Ai help a parent with this?

Athleet.Ai’s parent tools sit on the Parent plan, from £8.99 a month, and run across all eight sport editions, with 85 tools in total. They are built for exactly the problem above, which is that the useful signals are mundane and scattered across a child’s week. They help a parent keep a running total of organised sport hours across club, school and second sport, keep a dated record of where a pain started and whether it is persisting, see whether sleep is habitually short rather than occasionally late, and ask age-appropriate questions about load and rest. The Growth and Maturation Guide is there to explain what the growth spurt does and does not mean, which on the evidence above is mostly a reason to watch for pain at a growth area rather than to change training.

Four things they deliberately do not do. They do not predict injury, because no instrument reliably does that in a child. They do not produce a workload ratio and present it as a risk score, for the reasons set out above. They do not tell you that a weekly hours total is too high, because no such figure has been established, so the total is there to be seen, discussed and tracked over time rather than scored against a limit. And in My Fuelling, protected mode is always on for under-18s and hides calories and body weight, because a growing athlete does not need a number to fixate on. Where a tool’s answer is to see a doctor, that is the answer it gives.

Where the evidence stops

This literature is thinner than its confidence suggests. Most of it is cross-sectional or case-control, graded at the lower levels of evidence by its own authors, and the volume guidelines derive from two American hospital systems. The growth window rests on 26 players, the youth workload study on 32 who were 17, and the Osgood-Schlatter prognosis on 51 adolescents. The athletics-specific prospective evidence in this age band amounts to roughly one study of 101 Swedish children, in which 53% reported a new injury across a year and 91% of those injuries were non-traumatic (Ek, Kowalski and Jacobsson, 2022). The claim that 70% of children quit sport by 13 has no primary source anyone has been able to produce.

None of which means a parent should stop paying attention. It means the attention is better spent on the five signals that hold up than on the four numbers that do not.

Frequently asked questions

How many hours a week should a 12 year old train?

No evidence-based limit exists. The commonly quoted rule of no more hours a week than the child’s age in years lost statistical significance in the longitudinal study run by the group that proposed it, once total weekly sport hours were accounted for. What the evidence supports is that total hours of organised sport matter, with about a 31% increase in injury odds per additional five hours a week (1.31, 95% CI 1.19 to 1.44), and that a parent should count club, school and second-sport hours together rather than the club alone.

Is it bad for my child to specialise in one sport?

Two meta-analyses associate high specialisation with a higher injury risk, at pooled risk ratios of 1.81 and 1.37, though they rest on four and six overlapping primary studies and the mechanism is unknown. The clearer finding concerns performance. A meta-analysis of 6,096 athletes, including 772 of the world’s best, found that world-class adults had done more multi-sport practice as children and started their main sport later than their less successful peers, though the same analysis found that better junior performers had specialised earlier. Early specialisation tends to produce the stronger 14 year old. It is at senior level that the pattern reverses.

Should my child train less during a growth spurt?

Not on the current evidence. The largest study to measure growth directly found identical growth rates in injured and uninjured young athletes and concluded that growth rate was not related to injury risk. Persistent pain at a growth area such as the knee or heel is a different matter and should be seen by a GP. The reassurance that children always grow out of it may be too confident, since one small cohort study found that 37% of those followed up still had knee pain two years later. Most had recovered, and enough had not that persistent pain is worth an appointment rather than a wait.

How much sleep does a young athlete need?

The American Academy of Sleep Medicine recommends 9 to 12 hours a night for children aged 6 to 12 and 8 to 10 hours for 13 to 18 year olds. Habitually sleeping less than that carries a modest association with injury, pooled at 1.58 in a seven-study review. The evidence concerns chronic short sleep rather than a single late night, and the studies rely on self-reported sleep, so treat it as a pattern worth protecting rather than a precise dose.

Do training load apps predict injury in children?

No. The acute-to-chronic workload ratio, which most such products use, produced similar results when its chronic workload was replaced with randomly generated numbers, and did not outperform a model containing no information. The study usually cited as its youth evidence had a mean age of 17.3 years. No blood, hormonal or heart-rate-variability marker has shown the accuracy needed to flag overtraining in a young athlete either.

What are the signs my child is doing too much?

Five signs have evidence behind them, and they are a performance decline that does not recover with rest, persistent pain at a growth area, habitually short sleep, a problem the child is training through without mentioning it, and a high total of weekly organised sport hours. If anything about a young athlete’s health, energy or eating worries you, that is a GP appointment rather than a coaching question. Crossing a workload threshold, hitting the hours-equals-age number, being mid growth spurt, or seeming briefly clumsy are not reliable indicators on their own.

Nothing here is medical advice. Any persistent pain, or any concern about a child’s health, growth or eating, belongs with a GP.

The parent tools are available on the Parent plan across every Athleet.Ai edition. See what parents get or compare the plans.