The 10% Rule in Running: What Actually Predicts Injury
Patricia Bedoya
·
·
7
min read
Cover image generated with AI — © Chubby Studio S.L.
KEY TAKEAWAY
The 10% rule has almost no evidence behind it: in the only randomised trial, runners who followed it got injured as often as runners who did not. A 2025 study of 5,205 runners found a better warning sign, a single run much longer than your longest of the previous 30 days. After an injury that longest run has reset, so rebuild from your recent data and let your physiotherapist decide when you start.
The 10% rule says you should never increase your weekly running distance by more than 10%. It is the most repeated injury advice in running, and the best evidence we have does not support it. A randomised trial found the same injury rate with or without it, and a 2025 study of more than 5,000 runners found that one run much longer than usual was a better warning sign than any weekly total.
That does not make load irrelevant. Most running injuries still come down to asking a tendon, a bone or a muscle for more than it has been prepared for. It means the thing to watch is more specific than a weekly percentage, and that matters most when you are coming back from an injury and your sense of normal is out of date.
What is the 10% rule in running?
It is a rule of thumb: add no more than 10% to your total weekly distance compared with the week before. Run 20 km this week and next week's ceiling is 22 km. It is simple and it is everywhere, but it spread as coaching folklore rather than from a study, and a 2018 systematic review of training load and running injuries found no evidence supporting it.
Does the 10% rule actually prevent running injuries?
Not in the one randomised trial that tested it. In the GRONORUN study, 532 novice runners preparing for a 6.7 km event were split between a standard 8-week programme and a 13-week programme built on the 10% rule. Injuries hit 20.8% of the graded group and 20.3% of the standard one: five extra weeks of caution bought no protection at all.
The trial has limits. It studied beginners training for a short race, and both programmes were gentle by the standards of anyone chasing a half marathon. But it is still the only randomised test the rule has ever had, and the rule did not pass it.
Is there any safe weekly increase?
Nobody has found one. The closest result comes from 874 novice runners who were given GPS watches and followed for a year. Overall injury rates did not differ between those who increased weekly distance by under 10%, by 10 to 30%, or by more than 30%. For a specific group of overuse injuries, including runner's knee, shin splints and iliotibial band syndrome, those who increased by more than 30% over two weeks had a higher rate, but the result fell just short of statistical significance.
The authors called their study exploratory and suggested staying under 30% as a sensible margin rather than a proven limit. The 2018 review was blunter: very limited evidence links changes in training load, up or down, to injury at all. Weekly percentages are a weak lens, not a safety line.
See what each run really asks of you
FitWoody tracks your training load and analyses every run against your own cardiac load, alongside your sleep and recovery baselines, so you can see how much you are really asking of your body on the way back. Comes with a 7-day free trial.
Download the app
★★★★★
4.5 · 732 ratings on the App Store
What predicts running injuries better than weekly mileage?
The length of a single run compared with your recent longest run. In a 2025 study in the British Journal of Sports Medicine, Frandsen and colleagues followed 5,205 runners for 18 months and 588,071 sessions. Runs more than 10% longer than the longest run of the previous 30 days were linked to a higher injury rate, and the bigger the jump, the higher the risk.
Compared with runs that stayed within 10% of that recent longest, a run 10 to 30% longer carried a 64% higher injury rate, one 30 to 100% longer a 52% higher rate, and one more than double a 128% higher rate. The weekly measures told a different story: week-to-week changes showed no clear link, and the acute:chronic workload ratio pointed the opposite way from what the theory predicts.
It is an observational study, so it shows association rather than cause, and it cannot tell you that a run 9% longer is safe and one 11% longer is not. What it does suggest is that your body notices the one run far beyond anything it has recently done more than it notices a busy week built from several ordinary runs.
Why is a running injury a load problem at all?
Because tissue adapts to what it handles repeatedly and struggles with sudden jumps beyond that. Tendons, bone and muscle all get stronger in response to load, but slowly, over weeks and months. An overuse injury is what happens when the load arrives faster than that capacity grows.
The trap for runners is that the heart and lungs adapt much faster than the legs. A few good weeks can make a long run feel easy on the breath while the Achilles and the shins are still catching up, which is the mismatch we described in what ramping up too fast actually does.
Load is also more than distance. Hills, pace, a new surface or a new pair of shoes can each change what a familiar route costs your legs. And not every injury is a load problem: a twisted ankle on a kerb is bad luck, not bad planning.
How does this change coming back after an injury?
It moves the danger to the run where you feel fine. After weeks off, the longest run in your last 30 days is short or zero, so going straight back to your old long run is, by the measure in that study, a very large spike, even if it is a distance you used to cover every Sunday.
That is our reading of the research rather than something it tested, because the study followed runners in general, not people returning from injury. The logic still holds. Build the longest run back from what you have actually done in the last month, not from what you remember doing before you got hurt. When and how far you can start again is a question for the physiotherapist or doctor treating the injury, and pain during or after a run is something to bring back to them rather than something to push through.
If the time off was an illness rather than an injury, the same arithmetic applies to your weekly totals, and we wrote about which numbers to ignore after a sick week.
What can your watch tell you on the way back, and what can't it?
Your watch can show what each run asked of you: its distance and duration against your recent longest run, and its heart rate cost. It cannot show how a tendon or a bone is healing. Treat the data as a record of load, and your physiotherapist as the judge of whether the tissue is ready for it.
The useful parts are concrete. Your workout history gives you your longest run of the past 30 days, which is the number to stay close to. Heart rate shows how hard a run was on the inside, and the same easy run can cost you more on a bad day than on a good one. Resting heart rate and HRV against your own baseline show whether you are absorbing the work. None of them measures pain, stiffness the next morning, or how a healing injury is responding. Those are yours to notice and to report.
How FitWoody reads a comeback
FitWoody works from the data your Apple Watch records. It tracks your training load and analyses every run against your own cardiac load and your own heart rate zones, so a short, slow run after time off is read for what it cost you rather than flattened into a pace. Next to it sit your sleep and your recovery baselines, built from HRV, resting heart rate, temperature, blood oxygen and respiratory rate, so a comeback week that is quietly too much for you shows up in more than one place. For outdoor runs, Flyover replays the route with your heart rate moving along it.
Where to start this week
Look up the longest run you have done in the last 30 days and let that, not your old normal, set the length of your next long run. Keep any single run from jumping far beyond it. Stop treating a 10% weekly cap as protection, and stop treating a busier week as a failure when every run in it was ordinary. And if you are coming back from an injury, let the person treating it decide when you start, and let your recent data decide how far.
Frequently asked questions
Is the 10% rule in running a myth?
As a safety rule, largely yes. In the only randomised trial of it, novice runners on a 10%-a-week programme were injured at a rate of 20.8%, against 20.3% on a standard programme, and a 2018 systematic review found no evidence supporting it. Increasing gradually is still sensible. The specific 10% figure simply has no evidence behind it.
How much can I safely increase my long run?
No threshold has been proven. A 2025 study of 5,205 runners linked runs more than 10% longer than the longest run of the previous 30 days to a higher injury rate, and the risk grew with the size of the jump. Keeping each long run close to your recent longest is a reasonable guide, not a guarantee.
How do I avoid getting injured again when I return to running?
Follow the return plan from the physiotherapist or doctor treating the injury, because only they can judge how it is healing. On the load side, rebuild from what you have actually run in the last month rather than from your pre-injury routine, and avoid any single run far longer than your recent longest.