Skip to content
OpenAthlete

How Fast to Increase Running Mileage, Week by Week

The 10% rule did not reduce injuries when it was tested, and it compounds into nonsense. Steps, plateaus and planned lighter weeks work better, with one change at a time.

OpenAthlete Team5 min read

You have been running about 40 km a week for two months. Your marathon is 16 weeks away and you would like your biggest weeks to reach 70 km. This article is about getting from one number to the other without spending week 9 with a bag of ice on your shin. It is practical on purpose: for what the research does and does not say about load and injury, read running injuries and training load.

The 10% rule is a ceiling somebody made up

Nobody knows quite where the 10% rule came from, and it has been tested less than its fame suggests. When Dutch researchers gave 532 novice runners either a standard 8-week plan or a slower 13-week plan built on 10% increases, injury rates were the same: about one runner in five in each group (Buist et al., 2008). A Danish cohort of 874 novices found a possible signal only above 30% over two weeks, and even that was borderline (Nielsen et al., 2014).

The rule also fails in both directions. At low volume it is absurdly slow: 20 km a week becomes 22. At higher volume, applied every week without a break, it compounds. Forty kilometres plus 10% a week for ten weeks is 104 km. Nobody should do that, yet it is what the rule says.

What the evidence and the IOC consensus on load (Soligard et al., 2016) do support is the general idea: increase gradually, avoid sudden large jumps, and adjust to the person in front of you. The rest is coaching judgement, and I will label it as such.

How I build volume instead

Step, then hold. Add a chunk, say 4 to 6 km or 30 minutes, then stay at that level for a week or two before the next step. In my experience runners absorb a level they have repeated better than a new record every single week. That is observation, not trial data.

Lighter week every third or fourth. Cut volume by roughly a quarter to a third and keep one short quality session, for example 6 × 1' fast on the Wednesday, so you do not arrive at the next block feeling flat. The exact percentage has no strong study behind it. The habit of planning the week before you need it does.

One new thing at a time. Do not add 6 km in the same week as your first track session of the season, your first long hilly trail or a new pair of shoes with less drop. The Run Clever trial found that progressing intensity or progressing volume gave similar injury rates (Ramskov et al., 2018). Progressing both at once was not the question, and I would not volunteer to find out.

Grow the long run on its own track. Ten to fifteen minutes more each time it goes up, and early in a build keep it to about a third of the week or less. A runner on 40 km doing a 25 km long run is putting most of the week into one session.

Count what actually loads you. On the road, kilometres are a decent proxy. On trails, count time and elevation: 1,000 m of descent in a week is a stress that no distance figure captures, and your quadriceps will report it with a two-day delay.

Respect breaks. After a few days off, carry on. After two or three weeks off, I restart at half to two-thirds of the previous volume and rebuild over three or four weeks. Your recent average has dropped, even if your memory of it has not.

A 16-week example

Here is how the 40 to 70 km build could look. Long run durations assume an easy pace of around 6:00 per km.

WeekVolumeLong runNote
140 km1h20Current level
244 km1h30First step
344 km1h30Hold
434 km1h15Lighter week
548 km1h40
652 km1h45
752 km1h50Hold
840 km1h30Lighter week
956 km2h00
1060 km2h05
1160 km2h15Hold
1246 km1h40Lighter week
1365 km2h20
1470 km2h30Peak
1552 km1h45Taper
1630 km + raceRaceRace week

Look at week 5. Compared with the lighter week 4 it is a 41% jump, which would set off every alarm in a ratio-based app. Compared with weeks 2 and 3, the level the runner actually trained at, it is 9%. Compare a new week with the training before the recovery week, not with the recovery week itself.

Notice also what the plan does not do: it never adds volume and a new type of session in the same week, and the peak is two weeks out, not one. If your goal race is a hilly trail, swap kilometres for hours and add a column for D+.

What to compare a week against

If you want a single number, compare this week with the average of the previous three or four weeks, without counting this week in that average. Use it as a prompt to look more closely, not as a verdict. The popular acute:chronic workload ratio and its 0.8 to 1.3 "sweet spot" have been criticised heavily, including for that exact double-counting problem (Impellizzeri et al., 2020). There is no validated threshold, for you or anyone else.

When to hold or step back

Stop progressing, and usually stop running for a few days, if a pain changes the way you run, gets worse during a run, or is worse the next morning. Pain on one precise spot of a bone, like the shin or the top of the foot, deserves a professional opinion before the next run, not after the next three.

Hold the current level, or take the lighter week early, if easy runs feel harder than usual several days in a row, if the same easy pace costs you more heart rate, or if your sleep has gone. Those are the early signs covered in overtraining and overreaching. Moving a step one week later costs you almost nothing. A stress fracture costs you the race.

Planning it in OpenAthlete

The season view in OpenAthlete lets you draw a build like the one above as blocks (base, specific, taper, recovery) with your race on top, and each week shows its load against a range built from your recent weeks. That range is a rule of thumb to frame a week, not a safety limit. Fitness, fatigue and form are projected over the sessions you have planned, so you can see what a lighter week does to your form before you run it. Pain mentioned in the optional post-activity questions goes into an injury log your coach can see. More on the training calendar and training load pages.

Sources

  • Buist I 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. doi:10.1177/0363546507307505
  • Impellizzeri FM et al. Acute:chronic workload ratio: conceptual issues and fundamental pitfalls. Int J Sports Physiol Perform, 2020. doi:10.1123/ijspp.2019-0864
  • Nielsen RO et al. 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. doi:10.2519/jospt.2014.5164
  • Ramskov D et al. Progression in running intensity or running volume and the development of specific injuries in recreational runners: Run Clever, a randomized trial using competing risks. J Orthop Sports Phys Ther, 2018. doi:10.2519/jospt.2018.8062
  • Soligard T et al. How much is too much? (Part 1) International Olympic Committee consensus statement on load in sport and risk of injury. Br J Sports Med, 2016. doi:10.1136/bjsports-2016-096581

Written by

OpenAthlete Team

OpenAthlete is an open-source training platform built in Grenoble by an independent developer and ultra runner, with feedback from the coaches and athletes who use it. No investors, no ads.

Keep reading

Train with numbers you can check.

Every formula OpenAthlete uses is in its source code, and every feature is free on our servers or yours.

Free forever. No credit card.