RunFast
Running guides

How a marathon training plan is built

A marathon training plan combines rising weekly distance, a long run, easy and hard running, lighter weeks and a taper. The research backs some of these parts and says little about how to combine them.

By RunFast · Updated · 15 min read

This guide splits a marathon training plan into six parts: how far you run each week, how fast that distance grows, the long run, easy and hard running, lighter weeks and the taper. The research gives no recipe for putting them together: the authors of a 2024 analysis of 92 sub-elite marathon plans say scientific publications give little specific information for writing a comprehensive training plan.1 A 2020 review says there is no consensus on the best practice for recreational runners to train effectively, and the authors of a 2020 meta-regression say marathoners rely on expert opinion and their peers’ anecdotal advice when devising their plans.2, 3

What the research does support, for some parts, is an association: a 2020 meta-regression of 85 studies of elite and recreational marathon runners found that higher weekly distance and a longer longest run went with faster finish times.3 In a cohort of 441 recreational marathon runners, under 40 km (about 25 miles) a week and a longest run under 25 km (about 15.5 miles) went with slower finishes, and no training characteristic was linked to injury.4 In a training plan to run a marathon, that is an argument for building up weekly distance and the long run over the weeks.

What the research doesn’t support is a fixed weekly increase: a 2022 systematic review of 36 articles found the evidence linking injuries to weekly distance, or to changes in training, conflicting.5 One cohort does flag big jumps: in 735 runners training for the New York City Marathon, large increases in training volume were associated with more injuries.6 The taper and the mix of easy and hard running have guides of their own, on how long a marathon taper should be and 80/20 running and what the research says. If you are already running a marathon training plan from a book, a coach or an app, the checks below work on it too.

The parts of a plan, and what the research says about each

Here is what the research says about each part of a training plan for a marathon, and how sure it is. Each claim links to its source, and our editorial standards say how each one is checked.

Stages across the weeks

A 2022 systematic review of ten articles on highly trained and elite middle- and long-distance runners encourages a traditional periodisation pattern on a hard day-easy day basis.7 It describes their training intensity distribution, the mix of easy and hard running, shifting from pyramidal in the preparatory and precompetitive periods towards polarised in the competitive period.7 Read it as how such runners organise a season, not as a template for yours.

A 2024 analysis of the last 12 weeks of 92 sub-elite marathon plans set out what those plans recommend, and its authors note they have no evidence that the plans work.1

How much you run each week

The 2020 meta-regression pooled 137 cohorts of elite and recreational marathon runners, 25% of them female, from 85 observational and interventional studies.3 It looked at average weekly running distance, the number of runs a week, the most distance run in a single week, the distance of the longest run and hours of running a week.3 Each was associated with marathon performance, with higher values going with faster finish times.3

In the 2020 cohort of recreational runners, 441 marathon runners answered questionnaires on their average weekly training volume and their longest run.4 Under 40 km (about 25 miles) a week was related to a slower finish time, and over 65 km (about 40 miles) a week to a faster one.4

In the analysis of 92 sub-elite plans, recommended weekly volume over the final 12 weeks averaged 108 km, 59 km and 43 km (about 67, 37 and 27 miles) for the high, middle and low distance plans.1 Those are averages of what plans for sub-elite runners recommend, not targets that were tested.1

How fast to build

A 2022 systematic review of 36 articles found the evidence linking weekly distance, duration, frequency, intensity or changes in them with the onset of running injuries conflicting.5 It concluded that current evidence does not consistently link running injuries with specific training parameters or recent changes in them, and that caution should be taken when recommending optimal parameters or progressions.5

A 2018 systematic review, of runners aged 18 to 65, found four eligible articles, three of which linked increases in training load with a higher risk of running injury.8 It also reported no difference between a 10% and a 24% average increase in weekly volume, and concluded that very limited evidence supports a sudden change in training load being linked to injury.8 Its abstract gives the same hazard ratio as the novice cohort below for increases of over 30% against under 10% (1.59), so it probably draws on that study rather than confirming it.8, 10

A randomised trial in 532 novice runners preparing for a 6.7 km event tested a graded 13-week programme based on the 10% rule.9 Injury incidence was 20.8% in the graded group and 20.3% in the standard group, and the trial showed no effect of the graded programme on injury.9

A 1-year cohort of 874 novice runners with GPS watches grouped them by how their weekly distance changed: it rose by less than 10% or fell, rose by 10% to 30%, or rose by more than 30%.10 Injury rates did not differ significantly across the three groups.10 Its authors suggest novice runners may be well advised to increase their weekly distance by less than 30% a week over a 2-week period.10

A cohort followed 735 runners registered for the New York City Marathon (mean age 41.0, 46.0% female) through the 16 weeks before the race, with their training tracked on Strava.6 It compared each runner’s distance over the past 7 days with their distance over the past 28 days, a measure called the acute:chronic workload ratio, and found injury associated with the number of days on which that ratio was 1.5 or more.6 The authors conclude that increases in training volume of that size were associated with more injuries among runners training for a marathon.6

An International Olympic Committee consensus statement on load in sport says emerging evidence indicates that poor load management is a major risk factor for injury.11

The long run

In the meta-regression, the distance of the longest run was among the training behaviours associated with faster marathon finish times.3 In the cohort of recreational runners, a longest endurance run under 25 km (about 15.5 miles) was associated with a slower finish time.4 That cohort’s authors conclude that a high training volume and long endurance runs associate with a faster finish time but do not seem related to a higher injury risk.4

Easy and hard running

The review of elite runners encourages a hard day-easy day pattern, and in training logs from the 16 weeks before 151,813 marathons by 119,452 runners, the share of running in the lowest intensity zone was higher in progressively faster groups, an association rather than proof that it works.7, 12 For the mix in detail, see 80/20 running and what the research says.

Lighter weeks

A consensus statement on recovery in sport says an adequate balance between stress (training and competition load, and other life demands) and recovery is essential for athletes to achieve continuous high-level performance.13 In a plan, build in easier weeks or days at a rhythm that lets you recover.

The taper

A meta-analysis of competitive athletes found a 2-week taper with volume cut by 41–60% seemed the most efficient strategy, and one of endurance athletes found a taper of up to 21 days with the same cut effective, both without changing training intensity or frequency.14, 15 In more than 158,000 recreational marathon runners, tapers of up to 3 weeks were associated with better performance than shorter ones, and the guide to how long a marathon taper should be has the detail.16

How to apply it

This guide doesn’t give a week-by-week marathon training plan schedule. Instead, the first table gives you checks, part by part, for any training schedule for running a marathon, whether you found it in a marathon training guide or wrote it yourself.

Check a marathon plan: what to look for in each part, and what the evidence says
Part What to look for What the evidence says
Weekly distance A start near what you run now, then a build to the plan’s biggest weeks In recreational marathon runners, under 40 km a week went with slower finishes and over 65 km with faster ones, with no link to injury; plans for sub-elite runners averaged 43, 59 or 108 km a week in their last 12 weeks, with no evidence that they work1, 4
How fast it builds Steady rises, with no sudden big jump from one week to the next Two systematic reviews found the evidence linking changes in training to injury conflicting or very limited; in one marathon cohort, large increases in volume went with more injuries5, 6, 8
The long run A long run that grows over the weeks A longer longest run went with faster marathon times across 85 studies; in recreational marathoners, a longest run under 25 km went with slower finishes, and long runs didn’t seem related to injury risk3, 4
Hard and easy days Hard sessions with easier days between them A systematic review of elite runners encourages a hard day-easy day pattern7
Lighter weeks Easier weeks or days built in, at a rhythm that lets you recover A consensus statement calls a balance between stress and recovery essential for athletes13
The taper Two to three weeks of falling volume, keeping your faster running and your running days Two meta-analyses found volume cuts of 41–60% effective with intensity and frequency unchanged; in recreational marathoners, tapers of up to 3 weeks went with better performance than shorter ones14, 15, 16

A fixed weekly percentage also compounds. The second table shows what your starting weekly distance is multiplied by after 4, 8, 12 and 16 weeks of rising by the same percentage every week.

What a weekly increase adds up to: arithmetic on your starting weekly distance, not research
Weekly increase 4 weeks 8 weeks 12 weeks 16 weeks
6% 1.26 1.59 2.01 2.54
8% 1.36 1.85 2.52 3.43
10% 1.46 2.14 3.14 4.59

At 10% a week, 30 km (about 19 miles) would become about 138 km (about 86 miles) after 16 weeks, more than the 108 km a week that the highest-volume sub-elite plans averaged over their last 12 weeks.1 So a plan can’t keep adding a fixed percentage every week for months: look for a rise that levels off, with easier weeks along the way.

To set the pace of each kind of run, work out your training paces from a recent race with the pace calculator.

In RunFast

RunFast writes a marathon plan back from your race date, placing runs only on the days you can train. Onboarding asks for your current weekly distance, your longest recent run, your available training days and your goal. A race plan lasts 3 to 30 weeks and runs base, build, peak and taper into race day, and a marathon plan of five weeks or more tapers for three, race week included. Weekly distance starts from your current figure, never below 6 km, and rises 6% a week for beginners, 8% for novices, 9% for intermediate runners and 10% for advanced ones, up to a peak capped by your goal, level and starting distance. Every fourth week before the taper is a lighter week at about 75% of the week before, unless it is the last week before the taper. That weekly rise and that lighter-week rhythm are RunFast’s design choices, not results from a trial. One long run a week, at easy pace and on the weekend where your days allow, starts from your longest recent run (or 30% of your weekly distance, if that is more) and grows towards a cap of 34 km (about 21 miles) for a marathon, never more than 45% of the week’s distance. Up to two quality sessions a week, a tempo run at threshold pace or intervals, depend on stage and level, and the other days are easy runs, with very easy shakeouts after a hard day. Training paces come from Daniels’ VDOT model, worked out from a recent race or your fitness level.

Get RunFast on the App Store

Where the evidence is thin or mixed

No fixed weekly increase is proven

The randomised trial of the 10% rule was in novice runners preparing for a 6.7 km event, not a marathon, and it tested a whole 13-week programme built on the rule rather than the 10% figure alone.9 The cohort that compared rates of increase followed novice runners with GPS watches who were training on their own.10 In it, distance-related injuries were more common in runners whose distance rose by more than 30% than in those under 10% (hazard ratio 1.59, 95% confidence interval 0.96 to 2.66), but the difference was not statistically significant.10 Its authors’ advice, under 30% a week over 2 weeks for novice runners, is hedged and comes from that one cohort.10 Of the two systematic reviews, the 2022 one, of 36 articles, called the evidence conflicting, and the 2018 one, of four, called the evidence on sudden changes very limited.5, 8 In practice, treat a fixed weekly percentage as a rule of thumb rather than a tested limit.

Large jumps in load, and a debated ratio

The marathon finding is an association, from one cohort of runners who tracked their training on Strava.6 The ratio it used is itself debated: a 2020 critique says there is no evidence supporting the acute:chronic workload ratio in training-load management systems or in training recommendations aimed at reducing injury risk.17 The International Olympic Committee statement supports the broader point, that poor load management is a major risk factor for injury, but it is about elite sport and calls that evidence emerging.11 In practice, avoid sudden big jumps in weekly distance, without treating 1.5, or any other ratio, as a tested line.

Convention, not tests

The authors of the analysis of 92 plans note they have no evidence that the plans work, and the 2020 review says there is no consensus on the best practice for recreational runners to train effectively.1, 2 We found no study that tests how long each stage of a plan should last, how often to have a lighter week, or how many weeks to train for a marathon. So nothing here says a 16-week marathon training plan is better than a 12- or 20-week one: choose a length that fits your race date and lets you build steadily from where you are.

Associations, and who was studied

The meta-regression compared cohorts of elite and recreational marathon runners across 85 studies, so it shows which training went with faster times, not what caused them.3 The cohort of 441 recreational marathon runners rested on questionnaires they filled in themselves.4 The periodisation review drew on ten articles about highly trained and elite runners, and the analysis of 92 plans looked at what plans for sub-elite runners recommend, not at what runners did or gained.1, 7 Of the injury studies, the trial and the GPS-watch cohort were in novice runners, and the marathon cohort in New York City Marathon entrants who tracked their runs on Strava.6, 9, 10 In practice, use the numbers for direction and scale, not as targets.

Common questions

How many kilometres a week should a marathon plan reach?

The research gives associations, not a target: in recreational marathon runners, under 40 km (about 25 miles) a week was related to a slower finish time and over 65 km (about 40 miles) to a faster one, and no training characteristic was linked to injury.4 In plans written for sub-elite runners, recommended weekly volume over the last 12 weeks averaged 43 km, 59 km or 108 km, depending on the plan, with no evidence that the plans work.1 Build from what you run now, and treat these figures as context rather than a goal.

How long should the longest run be?

In recreational marathon runners, a longest run under 25 km (about 15.5 miles) was associated with a slower finish time, and long endurance runs did not seem related to a higher injury risk.4 Across 85 studies, a longer longest run went with faster marathon times.3 Take that as a reason to build your long run over the weeks, not as a distance to hit.

Is the 10% rule right for marathon training?

Not as a tested rule: in a randomised trial of 532 novice runners preparing for a 6.7 km event, a 13-week programme based on the 10% rule had no effect on injuries.9 One systematic review found the evidence linking changes in training to injury conflicting, and another reported no difference between 10% and 24% average weekly increases.5, 8 In a cohort of novice runners, probably among the second review’s studies, injury rates did not differ significantly between increases of under 10%, 10% to 30% and over 30%.8, 10 The trial and the cohort were in novice runners, not marathoners.9, 10 Rather than fixing on 10%, keep your rises steady and avoid big jumps.

Do I need a lighter week?

Some recovery, yes, and a lighter week, or recovery week, is one way to plan it: a consensus statement on recovery in sport says an adequate balance between stress (training and competition load, and other life demands) and recovery is essential for athletes to achieve continuous high-level performance.13 We found no study of how often a lighter week should come or how much lighter it should be, so pick a rhythm that lets you recover, and take an easier week when you need one.

This guide is general information, not medical advice. If you have an injury, a health condition or are pregnant, ask a doctor or physiotherapist before changing how you train or eat.

Sources

  1. Knopp M, Appelhans D, Schönfelder M, Seiler S, Wackerhage H. Quantitative Analysis of 92 12-Week Sub-elite Marathon Training Plans. Sports Medicine - Open. 2024. doi:10.1186/s40798-024-00717-5
  2. Boullosa D, Esteve-Lanao J, Casado A, Peyré-Tartaruga LA, Gomes da Rosa R, Del Coso J. Factors Affecting Training and Physical Performance in Recreational Endurance Runners. Sports. 2020. doi:10.3390/sports8030035
  3. Doherty C, Keogh A, Davenport J, Lawlor A, Smyth B, Caulfield B. An evaluation of the training determinants of marathon performance: A meta-analysis with meta-regression. Journal of Science and Medicine in Sport. 2020. doi:10.1016/j.jsams.2019.09.013
  4. Fokkema T, van Damme AADN, Fornerod MWJ, de Vos RJ, Bierma-Zeinstra SMA, van Middelkoop M. Training for a (half-)marathon: Training volume and longest endurance run related to performance and running injuries. Scandinavian Journal of Medicine & Science in Sports. 2020. doi:10.1111/sms.13725
  5. Fredette A, Roy JS, Perreault K, Dupuis F, Napier C, Esculier JF. The Association Between Running Injuries and Training Parameters: A Systematic Review. Journal of Athletic Training. 2022. doi:10.4085/1062-6050-0195.21
  6. Toresdahl BG, Metzl JD, Kinderknecht J, McElheny K, de Mille P, Quijano B, Fontana MA. Training patterns associated with injury in New York City Marathon runners. British Journal of Sports Medicine. 2023. doi:10.1136/bjsports-2022-105670
  7. Casado A, González-Mohíno F, González-Ravé JM, Foster C. Training Periodization, Methods, Intensity Distribution, and Volume in Highly Trained and Elite Distance Runners: A Systematic Review. International Journal of Sports Physiology and Performance. 2022. doi:10.1123/ijspp.2021-0435
  8. 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. doi:10.26603/ijspt20180931
  9. 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. doi:10.1177/0363546507307505
  10. Nielsen RØ, Parner ET, Nohr EA, Sørensen H, Lind M, Rasmussen S. Excessive progression in weekly running distance and risk of running-related injuries: an association which varies according to type of injury. Journal of Orthopaedic & Sports Physical Therapy. 2014. doi:10.2519/jospt.2014.5164
  11. Soligard T, Schwellnus M, Alonso JM, Bahr R, Clarsen B, Dijkstra HP, Gabbett T, Gleeson M, Hägglund M, Hutchinson MR, Janse van Rensburg C, Khan KM, Meeusen R, Orchard JW, Pluim BM, Raftery M, Budgett R, Engebretsen L. How much is too much? (Part 1) International Olympic Committee consensus statement on load in sport and risk of injury. British Journal of Sports Medicine. 2016. doi:10.1136/bjsports-2016-096581
  12. Muniz-Pumares D, Hunter B, Meyler S, Maunder E, Smyth B. The Training Intensity Distribution of Marathon Runners Across Performance Levels. Sports Medicine. 2025. doi:10.1007/s40279-024-02137-7
  13. Kellmann M, Bertollo M, Bosquet L, Brink M, Coutts AJ, Duffield R, Erlacher D, Halson SL, Hecksteden A, Heidari J, Kallus KW, Meeusen R, Mujika I, Robazza C, Skorski S, Venter R, Beckmann J. Recovery and Performance in Sport: Consensus Statement. International Journal of Sports Physiology and Performance. 2018. doi:10.1123/ijspp.2017-0759
  14. Bosquet L, Montpetit J, Arvisais D, Mujika I. Effects of tapering on performance: a meta-analysis. Medicine and Science in Sports and Exercise. 2007. doi:10.1249/mss.0b013e31806010e0
  15. Wang Z, Wang YT, Gao W, Zhong Y. Effects of tapering on performance in endurance athletes: A systematic review and meta-analysis. PLOS ONE. 2023. doi:10.1371/journal.pone.0282838
  16. Smyth B, Lawlor A. Longer disciplined tapers improve marathon performance for recreational runners. Frontiers in Sports and Active Living. 2021. doi:10.3389/fspor.2021.735220
  17. Impellizzeri FM, Tenan MS, Kempton T, Novak A, Coutts AJ. Acute:Chronic Workload Ratio: Conceptual Issues and Fundamental Pitfalls. International Journal of Sports Physiology and Performance. 2020. doi:10.1123/ijspp.2019-0864