How long does it take to train for an ultra?

The short answer

From a solid running base — you run several times a week, year round, and a half marathon does not frighten you — a first 50K takes four to six months of specific training. A 100 miler takes twelve to eighteen, and is better thought of as two or three seasons than as one training block.

From a standing start, add most of a year before that. Not to the training plan, but in front of it. Those months are for becoming a runner; the plan is for becoming an ultrarunner. Compressing them is the single most common way a first ultra ends in an injury rather than a finish.

Why your heart is not the constraint

Ask most runners what improves when they train and they describe the aerobic system: a stronger heart, more capillaries, more mitochondria, an easier breath at the same pace. All of that is real, and all of it arrives fast — a few weeks of consistent running is enough to feel it.

The structures that carry the load move on a different clock entirely. Bone turns over on a remodelling cycle lasting three to four months. When loading increases, osteoclasts resorb bone before osteoblasts rebuild it, so there is a window in which the bone is measurably weaker than when you started. If loading keeps rising through that window, microdamage accumulates faster than repair, and a stress reaction becomes a stress fracture.

Tendon is similarly slow. The systematic review of human tendon adaptation by Bohm and colleagues only admitted studies running eight weeks or longer, because shorter interventions do not reliably show change. Tendon responds well to load — magnitude matters more than volume — but it responds over months.

This is the gap that catches people. Ten weeks into a build you feel strong, your easy pace has dropped, and your training log says you can handle more. Your aerobic system is telling the truth. Your skeleton is three months behind it and has no way to say so.

What this means in practice

A twelve-week plan can make you fitter without making you ready. That is a perfectly good trade for a road marathon on a flat course, where the failure mode is a slow last 10 km. In a 100K on rough ground it is a different bet, because the failure mode is twelve hours of eccentric loading on descents that your connective tissue has not been given time to prepare for.

How fast should you build?

Here the honest answer is that the running world's favourite rule has no support behind it.

The 10% rule — never increase weekly mileage by more than a tenth — was tested directly. Buist and colleagues randomised 532 novice runners to either a thirteen-week graded programme applying the rule or a standard eight-week programme. Injury rates came out the same. A later systematic review by Damsted and colleagues examined the wider question and concluded that evidence for an association between sudden changes in training load and running injury is very limited.

That does not mean progression is safe at any rate. It means the number was invented rather than measured, and that a single percentage cannot capture what actually matters: how long you have been running, what your tissue has already adapted to, how much of the increase lands in one session versus spread across the week, and what the terrain does to the load.

The useful reframing is to stop asking how much you may add this week and start asking how many months the whole build has. If you have given yourself six months, you do not need a rule about weekly increments — there is room to progress gently and still arrive. If you have given yourself ten weeks, no rule will save you, because the arithmetic requires you to add faster than bone remodels.

Timelines by distance

Two starting points, because the same race is a different project depending on where you begin. "From a base" assumes you run consistently through the year and have raced a half marathon. "From scratch" assumes you are not currently a regular runner.

The pattern in the far-right column is worth reading carefully. As the distance grows, what limits you shifts away from anything a twelve-week block can install and toward things that only accumulate: descent tolerance, a gut trained to absorb carbohydrate while running, and the plain experience of having been out there a long time before.

Survey data points the same way. In a study of ultra-endurance mountain runners, those competing at distances over 90 km averaged 6.7 years of running experience, against 3.0 years for those at shorter distances, and reported around 10.7 hours of training per week. The study compared group means from self-reported questionnaires rather than modelling causation, so read it as a description of who lines up at those distances, not as a requirement. Still, the direction is consistent: the long races are populated by people who have been at it for years.

What counts as a base

"Build a base first" is advice everyone gives and almost nobody defines. A workable definition before starting an ultra build:

Planning across months, not weeks

If the timeline is measured in seasons, a plan has to survive contact with a season's worth of life: an illness, a work trip, a week where nothing goes right. A fixed spreadsheet written in January describes an athlete who no longer exists by March.

What matters is that the plan changes in response to what you actually did, and that it changes only forward. Weeks you have already trained are a record, not a draft. When you have a bad week, the question is what the next eight weeks should look like given that — not whether you can cram the missed sessions in somewhere.

Do it with other people

The other thing that survives a long build is company. A twelve-month project sustained purely on individual motivation is a hard bet; a Saturday long run that two other people are also showing up for is a much easier one. Training partners are worth choosing as deliberately as your shoes, and a shared plan means the long runs actually line up instead of drifting apart by a week.

Train 4 Ultra builds a plan across however many months you have before your race, sized to your current volume, your longest recent run and the days you can actually train. It adapts from your weekly check-ins and never rewrites the weeks behind you, and you can share a plan with the people you train with so you are all on the same long run.

Sources

• Bergman R, Kaiser K. Stress Reaction and Fractures. StatPearls Publishing, 2026. Bone remodelling cycle of three to four months, and the resorption-before-repair window in which bone is temporarily weaker.
• Bohm S, Mersmann F, Arampatzis A. Human tendon adaptation in response to mechanical loading: a systematic review and meta-analysis of exercise intervention studies on healthy adults. Sports Medicine – Open. 2015;1:7. Interventions of eight weeks or longer; loading magnitude as the key driver of tendon adaptation.
• Buist I, Bredeweg SW, van Mechelen W, Lemmink KA, Pepping GJ, Diercks RL. No effect of a graded training program on the number of running-related injuries in novice runners: a randomized controlled trial. The American Journal of Sports Medicine. 2008;36(1):33–39. 532 novice runners; no difference in injury rate.
• 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. International Journal of Sports Physical Therapy. 2018;13(6):931–942. Very limited evidence for the association.
• Belinchón-deMiguel P, Ruisoto P, Knechtle B, et al. Predictors of Athlete's Performance in Ultra-Endurance Mountain Races. International Journal of Environmental Research and Public Health. 2021;18(3):956. Group means from self-reported questionnaires; experience and weekly training hours by race distance.

This page is general information about training, not medical advice. If you have an injury, a health condition, or pain that persists, see a clinician rather than a training plan.