Peak-week mileage is the most misinterpreted number in marathon training. It is treated as a badge of seriousness when it is, in fact, a function of training history, current fitness, injury status, and goal time. The published evidence on marathon training volume points to a more modest conclusion than running culture suggests: peak weekly mileage should be the largest, but not by much, of a progressive build, and it should be reached only if the prior weeks have been completed without flag.
This article reviews what the research literature says about peak-week training volume for marathon preparation, what elite and amateur training patterns look like in practice, and how Indian runners — facing heat, air quality, and limited training-block calendar — should approach the question. The evidence is clearer than the folklore.
What the literature says about marathon training volume
A 2018 review in Sports Medicine by Tjelta examined training volume across elite male and female distance runners and reported peak weekly mileage typically falling between 160 and 220 km for international-class marathoners, with most training done at low to moderate intensity. This is the upper bound. It is not relevant to most amateur runners.
For amateur and recreational marathoners, the picture differs substantially. A 2012 prospective study published in the Journal of Sports Sciences tracked first-time marathon finishers and reported median peak weekly volumes of 50 to 70 km, with successful finishers (defined as completion without significant injury or excessive walking) clustering in the 55 to 65 km range. A separate 2014 retrospective survey in the British Journal of Sports Medicine found that amateur runners completing marathons in 4 to 5 hours typically peaked between 50 and 75 km per week.
The relationship between peak mileage and marathon time is not strictly linear above amateur ranges. Cumulative weekly running time, intensity distribution, and long-run length carry independent predictive weight. A runner reaching 70 km per week with consistent long runs and one weekly tempo session typically outperforms a runner reaching the same volume composed entirely of moderate efforts.
The dose-response question
The marginal benefit of additional weekly volume diminishes above thresholds specific to the runner. A 2016 meta-analysis in the European Journal of Applied Physiology reported that injury incidence rises measurably above 65 km per week in runners without an extended training history. Below that threshold, increased mileage tends to improve aerobic capacity and economy with limited injury cost. Above it, injury risk rises faster than performance gain.
This is why a 70 km peak is meaningful for one runner and excessive for another. The runner's two-year training history is a better predictor of safe peak mileage than the target time.
Realistic peak-week mileage by goal time
The following ranges synthesise the published literature and the prescriptions of widely-used marathon training systems including the Daniels VDOT, Pfitzinger, and Hanson approaches. They are not absolute prescriptions. They are reasoned defaults that account for typical training histories.
For a sub-3:30 marathon goal, peak weekly mileage between 65 and 85 km is broadly supported, with at least two quality sessions per week and a long run extending to 32 to 35 km. For sub-4:00, the supported range is 50 to 70 km, with one to two quality sessions and a long run of 28 to 32 km. For sub-4:30, 45 to 60 km is typical, with the long run extending to 26 to 30 km. For first-time finishers without a time target, peak weekly mileage of 40 to 55 km supports a safe completion when paired with a long run of 28 to 32 km.
Within these ranges, runners with longer training histories sit higher and runners coming off injury or with limited base training sit lower. The Daniels VDOT framework provides a more precise individual prescription, and the STRIDD calculators convert VDOT into weekly volume targets.
Long run as a fraction of weekly mileage
One consistent finding across the marathon training literature is that the long run should typically represent 25 to 35 percent of weekly volume. Long runs above 40 percent of weekly mileage correlate with elevated injury rates, per a 2017 retrospective review in Medicine and Science in Sports and Exercise. For a runner peaking at 60 km, this implies a long run of 18 to 21 km, with occasional outliers extending to 32 to 34 km in the final pre-taper weeks.
This is why peak-week mileage and long run length should be considered together. A 50 km peak with a 35 km long run is structurally different from a 70 km peak with the same 35 km long run, and the former carries more risk.
Indian conditions and peak-week realism
For most Indian runners, peak-week training occurs in October and November for a January marathon, or in November and December for a February marathon. These windows offer the only consistent block of moderately cool weather in much of the country, but they coincide with festival travel, Diwali disruption, and northern winter air quality concerns.
Realistic peak-week prescription in Indian conditions therefore requires a margin for unexpected disruption. A 2019 narrative review in the International Journal of Sports Physiology and Performance, summarising heat-acclimation and training-disruption literature, noted that peak training volume executed in compromised environmental conditions yields lower fitness adaptation than the same volume executed in tolerable conditions.
The practical implication is that Indian runners should not chase a target peak mileage if the conditions during peak week are unfavourable. Substituting one quality session per week with an indoor cross-training session, particularly during PM2.5 spikes in northern cities, preserves training stimulus without compounding environmental load. The types-of-run reference describes session structure in detail.
When to reduce peak-week mileage
Reduce the prescribed peak by 15 to 25 percent in the following situations, all of which are supported by the training-monitoring literature. Resting heart rate elevated by more than 5 beats above your morning baseline for three consecutive days. Sleep quality compromised for three of the seven prior nights. Any niggle that has persisted beyond two easy runs. Heat index exceeding 35 degrees Celsius during scheduled outdoor sessions.
The cost of executing a single hard week with a depleted body is typically not measured during that week, but in the two weeks that follow. A 2015 paper in the Scandinavian Journal of Medicine and Science in Sports identified the post-peak recovery period as the strongest predictor of taper-phase fitness retention. Cutting the peak intentionally is not weakness; it is calibration.
The taper, and what peak week actually predicts
Peak-week mileage is not, in itself, predictive of marathon performance. The published evidence consistently identifies cumulative training volume across the full block — typically the prior 12 to 18 weeks — as the stronger predictor, alongside the quality and consistency of long runs. A runner with one heroic peak week of 80 km but inconsistent training in the prior weeks generally underperforms a runner with steady weekly volumes in the 60 to 70 km range across the block.
The taper, which typically reduces volume by 30 to 50 percent across the final two to three weeks before race day, is supported by a 2007 meta-analysis in Medicine and Science in Sports and Exercise that reported small but consistent performance improvements with appropriate volume reduction. Maintaining intensity while reducing volume is the key principle.
For runners preparing structured plans for a January or February marathon in India, the STRIDD marathon plan sequences peak-week volume, long-run progression, and taper according to current fitness and goal time. The plan generator applies these principles individually, and the full Running Lab archive contains the underlying training-load methodology.
Peak week is the loudest week in a marathon block. It should be the most informed, not the most aggressive.