Sep 3, 2025
Risk and Reward: How much running is too much?
Introduction
Almost 10 years ago, Kaizen’s founder Josh would jog home from nights out and call it training. Some of his best races seemed to land the morning after pints with friends. Injury risk barely entered his head - stretching felt irrelevant, warm ups were optional - a classic approach in your early twenties. Nearing thirty, the pendulum has swung and like most he’s thinking differently about risk and reward in running.
The unbreakable body of a 20 something year old gets rusty as time goes on. Niggles and injuries creep up and before you know it, you’re out of action thanks to a bone stress injury, chronic knee pain, back pain or any other one of the dozens of injuries sidelining a passionate runner. Running injuries rarely occur thanks to some unfortunate accident (although this does happen) - they tend to be more common in field sports than in a sport that pretty much requires you to run in straight lines most of the time. But this brings it’s own issues, namely overuse injuries. And overuse tends to happen when you push your body beyond its current limits or capabilities.
In this article, we explore some new data to reinforce some well known theories on running injuries and a few new ones that might help any given runner avoid picking up lower extremity injuries.
Published in the British Journal of Sports Medicine, the authors conducted a study on a dataset of 5,205 runners and 588,071 logged runs over 18 months where injury risk tracked most strongly with how far a single run stretched beyond your recent long-run ceiling. But when chasing a PB, the practical message is to decide, consciously, where you sit on the risk–reward curve and to manage single-session spikes according to your goals, both short-term and long-term.
What the study did
Researchers followed 5,205 adult runners worldwide for 18 months. Participants wore Garmin devices and completed weekly injury questionnaires. The analysis looked at three ways a runner might “spike” load:
Single-session distance relative to the runner’s longest run in the previous 30 days
Acute:Chronic Workload Ratio comparing the latest week to the average of the prior three weeks (https://www.runnersworld.com/training/a20841513/acute-to-chronic-training-ratio-calculator/)
Simple week-to-week mileage ratio comparing one week to the immediately prior week
Injuries were self-reported as running-related overuse injuries. Mean age was 45.8 years and about 22% of the cohort were female.
The signal that mattered most
Compared with staying within 0 to 10% of your 30-day longest run, the hazard of injury rose as a single run stretched beyond that ceiling:
10 to 30% longer: hazard rate ratio 1.64
30 to 100% longer: HRR 1.52
More than 100% longer: HRR 2.28
Translated: push a single run notably farther than anything you have done in the past month and your injury risk rises quickly.
What did not help
Two popular week-level metrics underperformed in this cohort:
Acute:Chronic Workload Ratio (ACWR) showed a negative dose–response that did not align with injury risk. Remember: the ACWR compares the latest week to the average of the prior three weeks
Week-to-week mileage changes showed no relationship with injury.
An independent summary from Aarhus University, which led the work, emphasizes the same point and quantifies the cohort scale and risk jumps. About 35% of runners reported an injury during follow-up.
Important methods detail
The authors merged back-to-back activities if the pause between them was under 15 minutes. In other words, quick mid-run stops did not turn one long run into several “safer” short runs in the dataset.
Our takeaways
1) Pick your place on the risk–reward curve
Every training decision is effectively a trade-off between risk and reward. The study reinforces that the riskiest trade is a large jump in a single run. Where you sit on that curve should match your timeline.
Long-view development
Keep single-run distance within roughly 10% of your 30-day longest. Build total volume with frequency and doubles rather than by extending one run far past your ceiling. This keeps you on the flatter part of the risk curve while your tissues adapt.Short-term peaking
If a big race is looming and you accept higher risk for higher fitness, plan very selective nudges past that ceiling. One or two, separated by plenty of recovery, not a pattern of spikes. Expect risk to rise sharply as the overage grows. Do not string aggressive long runs together if your goal is longer term development. However, if your short-term goal is worth it, you might choose a higher risk for a higher reward.
2) No study is perfect. Recovery between runs still matters
This paper isolates a powerful predictor, but it does not make weekly load or recovery irrelevant. Two caution flags:
Compression of recovery windows
Splitting one very long run into two or three same-day runs may dodge a “single-session spike” label, yet it compresses recovery between sessions. That is still load your tissues must absorb. The IOC consensus on load and injury highlights how total load and inadequate recovery interact to raise risk. Same principle here.Short pauses do not change classification
Within this dataset, activities separated by less than 15 minutes counted as one session, so quick breaks did not fool the metric. If you truly split runs by hours, the paper does not show that risk is lower. Be careful claiming a “hack” where none was tested.
Bottom line for everyday programming: treat single-run distance, cumulative load, and recovery windows as connected levers. Use them together.
Strengths and limits worth knowing
Strengths
Huge sample, long follow-up, device-verified training distances, clear dose-response for a simple variable runners can monitor in real time.
Limits
Injuries were self-reported without clinical diagnoses. The sample skews to Garmin users and middle-aged recreational runners. The study interrogates distance; it does not fully isolate factors like pace, terrain, footwear changes, or concurrent strength work. That means external validity to different subgroups or mixed-modality plans needs caution. The authors also show ACWR and week-to-week ratios did not predict injury in this cohort, which challenges many watch and app heuristics used today.
Our conclusion
This study reframes injury risk around a simple, trackable behavior: how far you stretch a single run beyond what your body has proven it can handle lately. It does not erase the role of weekly volume, intensity, or recovery, but it shifts the primary warning light to the session level. The practical question becomes very simple. Where do you want to sit on the risk–reward curve today? Choose the lane that fits your horizon, then shape single runs, weekly distribution, and recovery around that choice. Any meaningful jump in fitness carries some risk. The art is deciding when to tap the brakes and when to press the gas, with eyes open to the trade.
Our cofounder Josh is a good example of a runner who’s willing to sit higher up the risk/reward curve as he prepares for a showdown with cofounder Michael in the Berlin marathon later this month. By pushing big 20km per week increases from 100km baseline weeks, he’s been pushing to get his fitness towards 2:28 shape in the hope of nailing a PB for the first time in over 8 years in the marathon distance. The risk is high but the payoff he’s aiming for is very clear. He went into this build with eyes wide open and awaits to see what the legs can do on race day.
Sources
Frandsen JSB et al. “How much running is too much? Identifying high-risk running sessions in a 5200-person cohort study.” Br J Sports Med. 2025;59(17):1203-1210. Key cohort details, HRRs, and null findings for ACWR and week-to-week ratios. PubMed
Aarhus University news release summarizing the cohort scale and risk increases, and contextualizing ACWR limitations. health.au.dk
BJSM article method note that sessions under 15 minutes apart were merged. British Journal of Sports Medicine
IOC consensus on training load and injury risk, emphasizing the interaction between load and recovery. Klokavskade
