How Much Should I Increase My Running Mileage Each Week?

The 10 percent rule is the most repeated piece of running advice there is: add no more than 10 percent to your weekly mileage. It is a reasonable default. It is not what the evidence says protects you. This guide covers what the research actually found, the rule Rebound Training uses instead, and a worked example so you can see what "gradual" looks like week by week.

Who this is for

Runners who are not currently dealing with an injury and want to build toward a race or a weekly mileage target without the classic mistake: a big jump, three good weeks, then a knee, shin or hip that ends the season. If you are coming back from a knee injury, start with the recovery guides instead; the rules there are stricter and the progression is a walk-run ladder, not a mileage build.

What the research says about the 10 percent rule

The one randomized trial found no protective effect. The GRONORUN trial randomized 532 novice runners to either a 13-week program built on the 10 percent rule or a standard 8-week program with average weekly increases of about 24 percent. Injury rates were 20.8 percent and 20.3 percent. Statistically identical (Buist et al. 2008).

A systematic review agreed. Reviewing the studies that measured training-load changes against running injuries, Damsted and colleagues found "very limited evidence" that a sudden change in load is associated with injury at all, and no support for 10 percent as a protective threshold (Damsted et al. 2018).

Big spikes do look different. In 874 novice runners tracked by GPS watch for a year, those who increased their weekly distance by more than 30 percent over two weeks had more distance-related injuries (patellofemoral pain, IT band syndrome, patellar tendinopathy, shin pain) than those who increased by less than 10 percent. The hazard ratio was 1.59 with a confidence interval that crossed 1, so it did not reach statistical significance, but the direction was consistent and the injuries were exactly the ones runners fear (Nielsen et al. 2014).

The single long run may matter more than the weekly total. A 2025 cohort of about 5,200 runners and 588,000 sessions from the same Aarhus group reframed the question around individual runs: a single run that exceeded the runner's longest run of the previous 30 days raised injury risk in that session, with runs 10 to 30 percent longer and runs more than 30 percent longer both showing a clear increase (Aarhus University 2025, press summary of the cohort).

Cutback weeks have a rationale, not a trial. No randomized trial shows that a lighter week every third or fourth week prevents injury. The mechanistic case is about bone: bone stress injuries typically surface three to four weeks after a jump in load, while remodeling bone is temporarily weaker, and periodic reductions give it time (Warden et al. 2014).

Order of progression is consensus, not proof. The Run Clever trial compared adding volume first against adding intensity first in 447 recreational runners and found no difference in overall injury risk, though the types of injury differed (Ramskov et al. 2018). "Change one thing at a time" survives as sensible practice rather than as an evidence-based law.

The honest summary: keep weekly jumps well under 30 percent, keep any single run close to your longest recent run, and take regular lighter weeks. Ten percent is a conservative way to do the first of those. It is not a guarantee.

The rules worth following

The rule, stated precisely

Rebound Training applies these rules mechanically, so it is worth saying exactly what they are.

• Each load week is capped at 10 percent above the previous load week. Not the previous calendar week: a cutback week is skipped when the next step is measured.
• Every third week is a cutback week at about 80 percent of the load week before it.
• The week after a cutback returns to the build line, which means it can be 40 to 50 percent above the cutback week. That is the saw-tooth shape a well-built plan has, and it is not a spike in the sense the research warns about: measured against the last load week, it is a normal step.
• Distances round to the nearest half mile or half kilometre, so the step you see can be a little larger or smaller than the raw percentage.
• The long run never grows more than about 10 percent past your longest run of the previous four weeks, and each race distance has a cap on the long run and on peak weekly volume.
• Volume only ever comes down in response to what you report. A high pain report (7 or more, or 4 or more at the shin or foot), two 4-to-6 reports in a row, or a run of missed sessions reduces the weeks ahead; a single 4-to-6 report softens the next quality session instead. Nothing accelerates a plan once it is running, and after a cut for pain or a red-flag pause, asking for more waits two weeks.
• The one exception is chosen up front. When the 10 percent rule cannot reach the long runs a race calls for before its date, Rebound may offer an accelerated build (up to 11 percent a week, long runs up to 15 percent) and asks you to confirm it when you activate the plan, beside a warning that it carries more injury risk. The evidence for that warning is the same cohort study above.

A worked example

A runner at 20 km a week, four run days, a 10k twelve weeks out and no recent race time. Rebound builds this:

Week 4 is 47 percent more than week 3. That is the point of the example: against the cutback week the step looks alarming, against week 2 it is a 9 percent step on the build line. Peak volume arrives in week 10 at a little over one and a half times the starting week, which is a twelve-week build without a single 30 percent jump between load weeks.

Stop and talk to a clinician if

• Pain that is sharp or stabbing rather than the dull ache of effort.
• Swelling anywhere in a joint.
• A knee or ankle that feels like it gives way.
• Pain at night or at rest.
• Pain that keeps building as you run instead of settling.
• One spot on a bone that hurts to press, or pain that changes your stride.

Rebound pauses a training plan on any of these reports and points you to a professional. That is a stop, not a judgment.

How Rebound fits in

Rebound Training builds your plan from where you are now: your current weekly mileage, your run days, your long-run day, and your race date or mileage target. The rules above are the engine's rules, applied to what the sessions actually add up to, not to a target on a spreadsheet. After every run you can log a 0 to 10 pain check and a short red-flag list. Pain of 4 to 6 holds the plan and softens your next hard session; 7 or more makes the next two runs easy and trims the volume ahead; a red flag pauses the plan. Volume only ever comes down in response to what you report. There is no morning check-in on a training plan and no goal-time input; the plan is built around your date and your days. In the installed iPhone app, Rebound can send today's workout to Apple Watch, and an Apple Health workout can prefill a log for you to confirm. Nothing logs itself.

Rebound is a fitness product, not medical care. This article is general education, not medical advice, and nothing here is a diagnosis. If something feels wrong, or a pain pattern does not settle, talk to a physical therapist or sports medicine clinician.

Sources

• Buist I, Bredeweg SW, van Mechelen W, et al. No effect of a graded training program on the number of running-related injuries in novice runners: a randomized controlled trial. Am J Sports Med. 2008;36(1):33-39. PubMed 17940147
• Damsted C, Glad S, Nielsen RO, Sørensen H, Malisoux L. Is there evidence for an association between changes in training load and running-related injuries? A systematic review. Int J Sports Phys Ther. 2018;13(6):931-942. PubMed 30534459
• Nielsen RØ, Parner ET, Nohr EA, Sørensen H, Lind M, Rasmussen S. Excessive progression in weekly running distance and risk of running-related injuries: an association which varies according to type of injury. J Orthop Sports Phys Ther. 2014;44(10):739-747. PubMed 25155475
• Aarhus University. Everything we thought about running injury development was wrong, study shows. 2025 press summary of the Garmin-RUNSAFE cohort. health.au.dk
• Ramskov D, Rasmussen S, Sørensen H, Parner ET, Lind M, Nielsen R. Run Clever: no difference in risk of injury when comparing progression in running volume and running intensity in recreational runners: a randomised trial. BMJ Open Sport Exerc Med. 2018;4(1):e000333. PubMed 29527322
• Warden SJ, Davis IS, Fredericson M. Management and prevention of bone stress injuries in long-distance runners. J Orthop Sports Phys Ther. 2014;44(10):749-765. PubMed 25103133