How Many Weeks to Train for a Marathon (and How to Split the Phases)

If you already run consistently, a marathon block takes 16 to 20 weeks. Runners who hold high mileage all year can complete a full build in 12 weeks; runners who are still logging low mileage need 24 weeks or more, because part of that time goes to building a base before the plan itself starts. The classic methods converge on that range: Hal Higdon and Hansons use 18 weeks, Pfitzinger offers 12 or 18, Jack Daniels considers 24 ideal, and the polarized plans in our catalog run 16.

The right number for you comes from three questions: how much you run today (weekly mileage and current long run), how many marathons you have finished, and what you want from the race, a finish or a time. Duration by itself does not make a plan work. What works is arriving at the specific phase with enough base to absorb race-pace workouts and 18- to 20-mile long runs.

Key takeaways

• 16 to 20 weeks is the standard for runners who run 3 to 4 days a week with a long run of 6 to 9 miles (10 to 15 km).
• 12 weeks only pays off if you have been at marathon volume (37 miles / 60 km a week or more) for months.
• A block has four phases: base, specific, peak and taper. The 2- to 3-week taper is the part you should cut last.
• In recreational runners, running under 25 miles (40 km) a week and never exceeding 15.5 miles (25 km) in the longest run were associated with slower marathons (Fokkema et al., 2020).
• Short on weeks? Cut from the base, never from the taper, and only if the base already exists.

How many weeks you need depends on where you start

Find the row in the table that describes your last 8 weeks, not the row you wish you were in.

The mileage you already run is the heaviest factor, and the research helps explain why. A study of 441 recreational marathoners in the Scandinavian Journal of Medicine & Science in Sports found slower finish times among runners who logged under 25 miles (40 km) a week and among those whose longest run never exceeded 15.5 miles (25 km). Those running more than 40 miles (65 km) a week finished faster (Fokkema et al., 2020). The same study found no association between those variables and injuries. The practical reading: the block needs enough weeks for you to spend time at those volumes, not to touch them once.

Experience shortens the timeline because the body already knows the load. A runner with two or three marathons behind them knows how long it takes to recover from a 20-miler and can handle a steeper ramp. Hal Higdon, for one, warns that his Intermediate 1 plan "is not recommended for runners doing their first marathon" (Hal Higdon, Intermediate 1).

The goal defines what fills the weeks. To finish, 18 weeks of an easy-mileage plan is enough. For a time goal such as a sub-4 marathon, the specific phase needs more weeks of race-pace work, and that pushes the whole block to 18 or more.

How long the classic methods run

The most followed methods agree on duration and disagree on content. This comparison uses each method's official sources.

Hal Higdon's Novice 1 runs 18 weeks with four runs at a comfortable pace, one cross-training day and two rest days. The long run climbs from 6 miles (10 km) in week 1 to 20 miles (32 km) in week 15, and the last three weeks are taper. For runners with little background, Higdon offers Novice Supreme, a 30-week program that adds 12 weeks of base training before the plan (Hal Higdon, marathon programs).

Pete Pfitzinger and Scott Douglas publish every Advanced Marathoning schedule in an 18-week and a 12-week version, across four mileage tiers: up to 55 miles a week, 55 to 70, 70 to 85 and over 85 (Human Kinetics). The 12-week version compresses the 18-week one rather than lightening it; in practice, it only fits runners who already hold the tier's entry mileage all year.

The Hansons method uses 18 weeks in both of its main plans and became known for capping the long run at 16 miles (26 km). The official Beginner plan starts at 12 miles in week 1, runs easy mileage only through week 5, and reaches 57.5 miles in week 15; the Advanced plan starts at 38 miles and reaches 61.5 (Hansons Running, official schedules). The logic is cumulative fatigue: the long run comes after a full week and makes up 25% to 30% of weekly volume. Luke Humphrey, the method's head coach, argues that "after 3 hours of running, you have crossed the point of diminishing returns" (Luke Humphrey Running). Note that Beginner is not a first-time-runner plan: it finishes at higher mileage than many intermediate plans.

Jack Daniels organizes any build into four phases and considers 24 weeks the ideal length, six per phase, opening with a six-week aerobic base; runners who already have that base can skip it and train for about 18 weeks (TrainingPeaks). The marathon chapter of Daniels' Running Formula includes an 18-week program for novices (Human Kinetics, book excerpt).

Polarized plans follow the distribution Stephen Seiler described in elite endurance athletes: roughly 80% of sessions at low intensity and 20% at high intensity, with little time in the middle zone (Seiler, 2010). They close in 16 weeks because the high easy volume is the base; the block only needs to insert and progress the hard sessions. The 80/20 polarized training guide shows how to lay out the week.

The four phases of a marathon block

Every method passes through the same four phases; what changes is where the borders sit. The split below is a starting point, and each plan in the catalog shifts a week or two one way or the other.

Base: arrive at the specific phase in one piece

The base exists so you can handle what comes next. It holds easy mileage rising week by week, a long run that grows to 12 to 15 miles (20 to 24 km) and short strides for mechanics, with no long hard sessions. The official plans show how modest this phase is. Novice 1's long run climbs from 6 to only 10 miles over the first five weeks; Hansons Beginner keeps easy running only through week 5, and quality workouts start in week 6. Runners who have held 30 miles (50 km) a week for months can shorten this phase; runners who have not turn it into the whole block, with no time left for the rest.

Specific: the plan becomes a marathon plan

This is where race pace enters the long run, threshold sessions appear, and the long runs reach 16 to 20 miles (26 to 32 km). It is the longest phase and the one that decides the result. In Hansons Beginner, the marathon-pace tempo run grows from 5 miles in week 6 to 10 miles in week 15. From week 11 on, the intervals switch from 5K-10K pace to "marathon pace minus 10 seconds per mile". Higdon's Intermediate 1 schedules two 20-milers, in weeks 13 and 15. The target pace for all of this comes from a recent race, not from a wish; run your latest half or 10K through the pace calculator and check your training zones before you start.

Peak: the two heaviest weeks, with nothing new

The peak holds the highest volume and the longest run of the build. Most 18-week plans place that run in week 15: 20 miles in Novice 1, and 16 miles inside a 57.5-mile week in Hansons Beginner. Nothing new here: shoes, gels and race strategy need to be tested on these long runs, because after them there is no other 18-mile opportunity.

Taper: cut volume, keep intensity

The taper is the planned reduction in volume over the final 2 to 3 weeks before the race. It is the phase with the most evidence behind it and the one runners sabotage most. A meta-analysis of 27 studies in competitive athletes concluded that the most efficient strategy was a 2-week taper with a 41% to 60% reduction in training volume while keeping intensity and frequency unchanged (Bosquet et al., 2007). In recreational marathoners, an analysis of more than 158,000 runners found that disciplined tapers of up to 3 weeks, with volume falling week after week, were associated with better races. A strict 3-week taper produced a median saving of 5 min 32 s (2.6%) compared with a minimal taper, and 64% of runners used less disciplined tapers (Smyth and Lawlor, 2021). The methods apply this differently. Higdon reserves the last three weeks; Hansons keeps its key workouts through week 17, with a 10-mile tempo at race pace ten days out, and only the final week is fully easy.

Before you start: what you need to have

Plans start from a point, not from zero. Week 1 of Novice 1 already has three 3-mile runs and a 6-mile long run. Week 1 of Hansons Beginner totals 12 miles in four runs; week 1 of Advanced, 38 miles in six. Pfitzinger's lightest schedule is described as "up to 55 miles a week". Luke Humphrey notes that "many times a runner is already running the weekly volume that the training plans start out at" and that the plans start a little easier by design (Luke Humphrey Running).

Those numbers come from the plans. The criteria below are coaching guidance, not science, and they work as a filter before you start a block:

1. Six months of consistent running with no active injury. Pain that changes your stride today becomes time off in week 10. The running injuries guide helps separate a niggle from a warning.
2. The ability to repeat week 1 of your chosen plan without being wrecked. If week 1 is already a jump from what you do now, you are short on base, and the answer is a longer plan, not a harder one.
3. A habitual long run of at least 90 minutes. Humphrey cites 90 minutes as a good introductory long run; at easy pace that is 8 to 9 miles (12 to 15 km).
4. A recent half marathon, ideally. It calibrates the pace for the specific phase and tests race logistics. Higdon suggests a half marathon in week 8 of Novice 1. If you have not run one yet, start with the half marathon plan.

What if you are short on weeks?

The honest question comes before any adjustment: does the base exist? If you have been running 30 miles (50 km) a week for months, 10 to 12 weeks make a useful block. If not, shortening the plan only concentrates the risk in the weeks that remain.

With the base in place, these practical rules apply:

• Never cut the taper. Two weeks is the minimum backed by the meta-analysis; three worked better in the recreational data.
• Cut from the base, not from the specific phase. It is the same logic Daniels applies when he lets runners with an aerobic base skip the foundation phase.
• Do not rush the long run. A safe progression is 1 to 2 miles a week, with a cutback week every third or fourth week. If the weeks available do not allow you to reach 17 or 18 miles at that rate, what changes is the time goal, not the progression.
• Trade the goal, not your health. "Finish" instead of "sub-4" is a coaching decision; "sub-4 no matter what" with 8 weeks and no base is a decision week 6 collects on.

The 8-week plan below fits that case: four running days built on the Daniels method, designed for runners who pair running with strength training and arrive with the base ready.

Mistakes that shorten the block without you noticing

Starting week 1 without being able to run week 1. The runner picks the plan by the method's name rather than by their own mileage and spends the first four weeks in deficit. The fix is to choose the plan whose week 1 looks easy.

Switching methods mid-block. Each plan distributes the phases differently; switching in week 9 usually means repeating one phase and losing another. If the plan feels too heavy, reduce volume inside it instead of jumping to another.

Stretching the long run and forgetting the rest of the week. In the Hansons plans the long run is 25% to 30% of weekly volume (Luke Humphrey Running), and the rest of the week is what carries the race. A 20-miler inside a 28-mile week is an isolated workout, not a preparation.

Frequently asked questions

Can you train for a marathon in 12 weeks?

Yes, if you have been running 37 miles (60 km) a week or more for months and have at least one marathon behind you. Pfitzinger and Douglas publish 12-week versions of all their schedules, and Daniels allows runners who already have a base to skip the six-week foundation phase. Without that base, 12 weeks make a good half marathon build, not a marathon build.

Do I need to run 20 miles before the race?

Not necessarily. Hansons caps the long run at 16 miles inside weeks of 55 miles or more, while Higdon's Novice 1 reaches 20 miles inside much smaller weeks. What the recreational data associates with slower races is never exceeding 15.5 miles (25 km) in the longest run (Fokkema et al., 2020). Beyond that, the long run's length depends on the weekly volume it sits in.

How many weeks should the taper be?

Two to three. The Bosquet et al. (2007) meta-analysis pointed to 2 weeks with a 41% to 60% volume reduction as the most efficient strategy. In Smyth and Lawlor's (2021) recreational data, the disciplined 3-week taper paid off most.

18 or 24 weeks: which should I choose?

Choose 24 when the base has to be built inside the block, meaning you cannot yet run week 1 of an 18-week plan comfortably. Choose 18 when you already run the entry mileage. To decide between methods, the marathon training guide compares a typical week of each; the plan catalog lists the length and volume of every option.

References

• 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;30(9):1692-1704. https://doi.org/10.1111/sms.13725
• Bosquet L, Montpetit J, Arvisais D, Mujika I. Effects of tapering on performance: a meta-analysis. Medicine & Science in Sports & Exercise, 2007;39(8):1358-1365. https://doi.org/10.1249/mss.0b013e31806010e0
• Smyth B, Lawlor A. Longer disciplined tapers improve marathon performance for recreational runners. Frontiers in Sports and Active Living, 2021;3:735220. https://doi.org/10.3389/fspor.2021.735220
• Seiler S. What is best practice for training intensity and duration distribution in endurance athletes? International Journal of Sports Physiology and Performance, 2010;5(3):276-291. https://doi.org/10.1123/ijspp.5.3.276
• Hal Higdon. Novice 1 Marathon Training Program. https://www.halhigdon.com/training-programs/marathon-training/novice-1-marathon/
• Hal Higdon. Intermediate 1 Marathon Training Program. https://www.halhigdon.com/training-programs/marathon-training/intermediate-1-marathon/
• Hal Higdon. Marathon Training Programs (overview, including the 30-week Novice Supreme). https://www.halhigdon.com/training/marathon-training/
• Pfitzinger P, Douglas S. Advanced Marathoning, 3rd ed. Human Kinetics. https://us.humankinetics.com/products/Advanced-Marathoning-3rd-Edition
• Hansons Running. Training Plans (official Beginner Marathon and Advanced Marathon schedules, Hansons Coaching Services). https://hansons-running.com/pages/training-plans
• Humphrey L. Hansons Marathon Method: The 16 Miler. https://lukehumphreyrunning.com/hansons-marathon-method-the-16-miler/
• Humphrey L. Hansons Marathon Method: should I reduce my mileage at the beginning? https://lukehumphreyrunning.com/hansons-marathon-method-should-i-reduce-my-mileage-at-the-beginning/
• TrainingPeaks. How Long Should My Running Training Cycles Be? https://www.trainingpeaks.com/blog/how-long-should-my-training-cycles-be/
• Human Kinetics. Sample running program from Daniels' Running Formula (book excerpt). https://us.humankinetics.com/blogs/excerpt/sample-running-program-from-daniels-running-formula