Marathon Pace KM
Menu

20 miles or three hours? Choosing your longest marathon run

Twenty miles and three hours are different limits for different runners. At 5:00/km, 32 km takes 2:40. At 7:00/km, it takes 3:44, before stops. Choosing a longest run from the distance alone can hide a very different training day.

Compare the distance, likely duration, recent long runs and the rest of your week. Then give the session a purpose you can rehearse. A longer number on the watch is useful only if it fits the preparation around it.

Does your longest run have to be 20 miles?

No single distance is a compulsory entry test for a marathon. Twenty miles is about 32.19 km, a familiar landmark in many plans, but it is not a universal requirement. The B.A.A.’s official Level One plan description, for example, sets its longest runs at 16–18 miles within an established weekly programme. The surrounding preparation matters.

That does not mean any shorter run is automatically enough. It means choosing a whole plan, rather than extracting the longest Sunday from another runner’s calendar. Consider the running frequency, ordinary weekly distance, long-run progression, intensity and recovery the plan assumes.

Three hours can be a useful coaching boundary, especially when a fixed distance would turn into a much longer outing. It is not a biological switch where useful training ends at 2:59 and becomes harmful at 3:01. A Marine Corps Marathon coaching article discusses a 3–3.5-hour limit and acknowledges differing coaching approaches. That is practical guidance, rather than proof of one optimal cap for everyone.

If you expect a longer marathon, the answer is not to run the entire race distance in training to remove uncertainty. Discuss the long-run budget and wider build with a coach who can consider your background. If the opening workload of your chosen plan is already a stretch, start with the training-duration and entry-workload guide.

Compare the run with your week and your recent longest outing

Enter a candidate long run, your realistic average moving pace, planned stopped time and the total running distance for that week. The weekly total must already include the long run. Add your longest run from the previous 30 days to make the proposed distance change visible.

Worked example: 30 km at 6:30/km takes 3:15 moving time. Ten minutes stopped makes 3:25 elapsed. In a 60 km week, that one run is 50% of the distance and leaves 30 km for all other runs. Compared with a recent longest run of 28 km, it is 7.1% farther. None of those percentages establishes that the session is suitable.

Compare another week: the same 30 km is 60% of a 50 km week, or 37.5% of an 80 km week. Those are different distributions. Raising the weekly total just to improve the percentage would add more work; a nicer-looking ratio is not a reason to do it.

What does the evidence say about the longest run?

Fokkema and colleagues’ 2020 study included 441 marathon trainees. Longer runs and higher weekly distance were associated with faster finishes, but runs beyond 35 km were not associated with better performance than the 30–35 km category. This was observational, with self-reported training and relatively few runners in the highest-distance category. It cannot establish that adding kilometres causes a faster marathon or identify your ideal duration.

The study also did not identify an association between its training categories and injuries. That does not establish that every long-run increase is safe. Categories of typical training and abrupt changes in an individual session are different questions.

A 2025 cohort study of 5,205 runners found higher overuse-injury rates when a session exceeded the runner’s longest distance in the preceding 30 days by more than 10%. It used device-recorded distance and self-reported injuries. The finding concerns associations, not an individual injury forecast, and the reference group was not injury-free. Staying beneath 10% is not a safety guarantee or an instruction to increase every week.

Together, these studies support looking at the run in context. They do not establish a universal three-hour ceiling, a mandatory 32 km rehearsal or a rule that 30% of weekly distance is always the right share. The worksheet intentionally shows the quantities without turning them into a clearance badge.

Choose the next run using four checks

1. What has the recent progression actually been?

Write down the last few long runs, with distance, elapsed time and how the following days went. If the last month’s longest was 24 km, moving directly to 32 km is a 33.3% distance increase. A calendar saying “peak week” does not make that history disappear. Consider repeating or extending more modestly with the person guiding your plan.

2. What is this run meant to teach you?

A rehearsal might test shoes, clothing, breakfast timing, carrying fuel and finding water. Those tasks deserve an explicit plan. You do not need to add a fast finish merely because it is the longest outing. For intensity choices within an established distance, use the separate long-run structure guide.

3. What else is in the week?

Check the previous hard session, total distance, work commitments and the days afterwards. A long run with marathon-pace sections is a different workload from covering the same distance easily. The monthly planner can help lay out a week, but its generated numbers do not supersede current recovery or an individual plan.

4. What would make you shorten it?

Choose accessible water, a route with an exit and a way home. Shorten or stop if pain changes your running or conditions make the plan unsuitable. If illness, injury or repeated fatigue has interrupted the build, use the disrupted-training guide rather than trying to rescue the calendar with one oversized session.

What my 32 km rehearsal actually taught me

On 9 August 2026 I ran 32 km with early two-kilometre laps around 5:16–5:31/km. I was using a thick homemade carbohydrate mix, had stomach cramping around 21 km and a stitch around 25 km, and stopped for water and carbohydrate. I also needed commercial gels. The full fueling account explains what I recorded and what remains uncertain.

The lesson was not simply that I had “ticked off 32”. Pace and the practical difficulty of taking fuel occurred together. I did not isolate their effects, and I cannot use one uncomfortable rehearsal to prove that 32 km was too long or that one ingredient caused the problem.

For the next rehearsal, a useful question would be whether I can execute a more controlled effort and a clearly recorded intake plan. That is a more informative aim than repeating the distance solely to remove a doubt. Your useful lesson may be different, but it should be something you can describe afterwards.

When should the final longest run happen?

Place it within the plan’s taper, rather than adding it wherever a free Sunday appears. The two- and three-week taper worksheet compares the distance budgets and keeps race distance separate. The right timing also depends on the session’s intensity, your response and the rest of the build.

A missed longest run late in training does not automatically belong in race week. Count the completed preparation and adjust expectations. Extra reassurance is a poor reason to add a session whose recovery you cannot observe before the event.

Long-run debrief

Purpose / distance / moving and elapsed times: ____________________

Effort and conditions compared with the plan: _____________________

Food and drink planned / actually consumed: ______________________

Any discomfort, route changes or stops: __________________________

How subsequent easy running felt, and next decision: ______________

Judge the run by what it contributes to the build: a manageable training day, a clearer race routine and information for the next decision. The milestone distance is only one line in that record.

Sources and calculation notes