All Articles

For decades, runners have been told to worry about their weekly mileage. Don't increase too fast. Follow the 10% rule. Keep your weekly totals in check.

But a 2025 study published in the British Journal of Sports Medicine just rewrote the playbook. After tracking 5,205 runners across 588,000+ individual sessions over 18 months, researchers found something surprising: weekly mileage changes were a weak predictor of injury. The much stronger signal? How far you ran in a single session compared to your recent history.

This isn't a minor tweak to existing advice. It's a fundamental shift in how we should think about injury prevention.

What Did the BJSM Study Actually Find?

"When a single running session exceeded 10% of the longest run in the past 30 days, overuse injury rates increased significantly. Small spikes raised risk by 64%. Large spikes (more than doubling your longest recent run) raised it by 128%."

— Frandsen et al. (2025), British Journal of Sports Medicine, n=5,205

The study, led by researchers at Aarhus University, collected Garmin data from runners in 87 countries. Participants averaged 46 years old, 78% were men, and most had 4–20 years of running experience. Of the 5,205 runners, 1,820 (35%) sustained a running-related injury during the study period.

Researchers defined three types of "spikes" based on how much a single run exceeded the runner's longest session in the previous 30 days. Then they measured which metric best predicted injuries.

588,000+ individual running sessions analyzed across 87 countries over 18 months, making this one of the largest running injury datasets ever assembled

The Spike Categories and Their Risk Levels

The researchers defined session spikes relative to each runner's longest run in the past 30 days. Here's how the risk broke down:

Spike Category Definition Injury Risk Increase
Baseline Run within 10% of longest recent run Baseline (reference)
Small spike 10–30% above longest recent run +64%
Moderate spike 30–100% above longest recent run +52%
Large spike More than double longest recent run +128%

The pattern is clear. Once you push a single run beyond 10% of your recent longest, injury risk climbs. And the further you push, the steeper the climb.

What's notable is that even "small" spikes (10–30% above your recent max) carry meaningful risk. This isn't just about avoiding dramatic one-off efforts. Incremental overreach in individual sessions adds up.

Why Didn't Weekly Mileage Predict Injuries?

"The study found no significant relationship between week-to-week mileage changes and injury. The traditional focus on weekly volume progression may be misplaced. Individual run distance relative to recent training load matters more than weekly totals."

This is the finding that challenges decades of running advice. When the researchers tested week-to-week mileage ratios as injury predictors, the signal was weak. The data simply didn't support the idea that a 15% or 20% jump in weekly volume was the primary driver of injury.

Why? Think about how weekly mileage works in practice. A runner could add 5 miles to their week by tacking 1 mile onto each of their 5 runs. Or they could add those 5 miles to a single long run. The weekly total is identical. The injury risk is not.

The study also tested the Acute:Chronic Workload Ratio (ACWR) and found a negative dose-response relationship, meaning it didn't predict injuries in the expected direction either. The session-level metric outperformed both established approaches.

What Does This Mean for How You Train?

"Instead of obsessing over weekly totals, focus on protecting individual sessions. No single run should exceed 110% of your longest run from the past 30 days. When you want to add volume, spread it across multiple easy runs rather than loading it into one session."

The practical implications are straightforward:

  1. Track your longest run, not just your weekly total. Before any run that's going to be longer than usual, check: is this more than 110% of my longest run in the past month? If yes, cap it there.
  2. Add mileage through frequency, not distance. Instead of making your long run longer, add an extra easy run to your week. Five 4-mile runs is safer than three 4-mile runs plus one 8-mile run (if your recent max was 5 miles).
  3. Grow your long run incrementally. When you do want to push your long run further, increase by no more than 10% of your current longest. Then hold that distance for 2–3 weeks before pushing again.
  4. Be especially careful after time off. If you missed a week or two, your "longest recent run" has reset. Don't jump back to your pre-break long run distance on day one.

How This Connects to the Bigger Picture

This study doesn't exist in isolation. It adds to a growing body of evidence that running injuries are more about individual session management than aggregate volume.

Jack Daniels' equilibrium method already advocated for step-loading: increase your volume, then hold for 3–4 weeks before increasing again. The BJSM study explains why this works. Holding your distance steady gives your body time to establish a new baseline, which means future runs at that distance no longer register as spikes.

The 3-up-1-down pattern (build for three weeks, deload on the fourth) achieves the same thing. Each build week pushes the long run slightly further, but the deload week prevents cumulative spike fatigue.

Research on the Acute:Chronic Workload Ratio (keeping your ACWR between 0.8 and 1.3) also aligns. The ACWR sweet spot essentially describes training that avoids large spikes in any direction.

What Should Running Apps Be Doing Differently?

Most training apps focus on weekly mileage caps. They'll warn you if your week is trending above plan. But very few monitor individual session distances against your recent history.

Based on this study, a well-designed training plan should include a session-level spike guard that prevents any single run from exceeding 110% of your longest run in the past 30 days, regardless of what your weekly total looks like.

It should also distribute volume increases across multiple sessions rather than concentrating them in one long run. And it should be especially protective after gaps in training, when your recent-run baseline has dropped.

After returning from time off, your "longest recent run" baseline will have dropped. Resist the urge to pick up where you left off. Your body's tolerance has reset, even if your fitness hasn't fully decayed.

Key Takeaways

  • Single-session distance spikes predict running injuries far better than week-to-week mileage changes
  • Exceeding 110% of your longest run from the past 30 days significantly increases overuse injury risk
  • Small spikes (+10–30%) raise risk by 64%; large spikes (doubling your longest run) raise it by 128%
  • Weekly mileage ratios showed no significant injury prediction in this study of 5,205 runners
  • Add volume through more frequent easy runs, not longer individual sessions
  • After time off, rebuild gradually; your recent-run baseline has reset
  • Hold new long-run distances for 2–3 weeks before pushing further

Pheidi's session-level spike guard protects every run

Your plan automatically caps individual run distances based on your recent training history. No single run exceeds 110% of your longest recent session, so you build distance safely without doing the math yourself.

Get Your Free Plan

References