Inside elite marathon training: what it actually takes to run a sub-2 hour marathon

A 1:59:30 marathon is not a lightly faster version of a 2:30 marathon. It is a different physiological event. To run 42.195 kilometres at 2:50/km — roughly 21 km/h, sustained for nearly two hours — every system in the body has to be tuned to a level that is, statistically, about one-in-a-billion. Engine size, fuel efficiency, gait economy, heat dissipation, mental tolerance for sustained discomfort: all of them have to land in the right place at the right time, in the right shoes, on the right day.

This piece is about the underlying biology and training that produces that combination. It is grounded in what we know, in 2026, about Sabastian Sawe, Eliud Kipchoge, the late Kelvin Kiptum, and the small number of men's and women's marathoners who have come within striking distance of two hours. For the race that finally broke the barrier, see how Sabastian Sawe broke the 2-hour marathon. For the man behind it, see our profile of Sawe. This piece focuses on the engine room.

The three numbers that matter most

If you boil elite marathon physiology down to its essentials, three numbers do most of the work.

VO₂max — the size of the engine

VO₂max is the maximum rate at which the body can take in, transport and use oxygen. It is measured in millilitres of oxygen per kilogram of body weight per minute (ml/kg/min). It correlates loosely with raw aerobic potential — bigger engine, more sustainable power.

Untrained adult male: ~35–45 ml/kg/min.
Recreational marathoner (3:30–4:00): ~50–55 ml/kg/min.
Sub-3 marathoner: ~60–65 ml/kg/min.
National-class marathoner (2:15-ish): ~70 ml/kg/min.
World-class sub-2:05 marathoner: 71 ± 6 ml/kg/min in the most recent published study of the cohort.
Theoretical ceiling: values in the high 80s and low 90s have been measured in a small number of cross-country skiers and cyclists. The marathon-specific ceiling appears to be lower because of the impact load of running.

The interesting part is what is not on this list: an absurdly high VO₂max. Sub-2 marathoners are not the highest-VO₂max athletes on Earth. They are athletes with very high VO₂max numbers combined with two other things — efficient running economy and an unusually high fractional utilisation. Everything below is about how those three pieces interact.

Lactate threshold — the size of the cruising window

If VO₂max is the engine, lactate threshold is the cruising speed. It is the highest sustained intensity at which lactate produced by the working muscles is matched by the rate at which the body clears it; above that threshold, lactate accumulates, pH drops, and the runner is on a clock. In trained runners it is usually expressed as a percentage of VO₂max, or as a pace.

Recreational runner: lactate threshold sits around 75–80% of VO₂max.
Well-trained sub-elite: 80–85% of VO₂max.
World-class sub-2:05 runner: lactate threshold around 83% of VO₂peak, with the lactate turn-point (the second, harder threshold) at 92% of VO₂peak.

The implication is direct. A sub-2 runner is comfortable running at 83% of their maximum aerobic capacity, and can hold close to 92% for two hours. A typical recreational marathoner, running their goal pace, is somewhere around 80–85% — and is usually tipping over their threshold by 30km, which is why marathons hurt the way they do at the back end. The elite has a wider cruise control band and a higher ceiling on it.

Running economy — how much fuel the engine uses

The least visible of the three, and arguably the most important. Running economy is the metabolic cost of running at a given pace — typically expressed as oxygen consumption (ml/kg/min) at a fixed sub-maximal speed. Two runners with identical VO₂max can have very different economies, and the more economical one will be faster at race pace.

Elite Kenyan marathoners have repeatedly tested with some of the lowest oxygen costs ever measured. The proposed reasons are a mix of biomechanical (long Achilles tendons, slim lower legs that swing easily, narrow pelvis-to-hip ratio that reduces side-to-side waste), neuromuscular (a high proportion of slow-twitch fibres, an efficient stretch-reflex), and developmental (decades of running from childhood at altitude, on uneven surfaces, before any structured training begins).

In the published literature, sub-2:05 marathoners run at about 190–200 ml/kg/km of oxygen cost — roughly 10–15% better than typical sub-elites. That difference is, by itself, the difference between 2:05 and 2:00.

The physiology of sub-2 pace, in one paragraph

A 2026 review of athletes capable of running at sub-2 marathon pace put the demand around 75–85% of VO₂max sustained for two hours, on legs that are taking somewhere between 170 and 190 strides per minute with a contact time around 150–180 milliseconds, fuelled by glycogen plus 60–90 grams of carbohydrate per hour, in shoes that improve running economy by another 3–5%. Of the elite athletes in the lab who were tested at exactly 2-hour marathon pace on a treadmill, only seven of sixteen were able to reach a steady aerobic state at that intensity. The other nine were drifting into oxygen debt and would not have lasted the distance. That is the population from which the actual sub-2 runners are drawn.

Weekly mileage: what they actually run

There is a stubborn myth that elite marathoners log a more-or-less standard mileage. They don't. The published training of the modern sub-2:01 cohort spans a wide range, and the differences are part of what makes the era interesting.

Eliud Kipchoge has, for most of his career, trained at 180–220 kilometres per week (110–135 miles), with relatively conservative high-end intensity and an unusual emphasis on consistency. His coach, Patrick Sang, has spoken publicly about resisting the temptation to add volume.
Kelvin Kiptum trained at 250–280 kilometres per week, sometimes pushing past 300km. His training included two long runs of 40km at marathon pace per week — Thursdays and Sundays — a load that other elite coaches described as both extraordinary and concerning. His coach, Gervais Hakizimana, said publicly that he worried about injury.
Sabastian Sawe, in the build to London 2026, averaged ~200 km per week in the six weeks before the race — close to Kipchoge's range, well below Kiptum's.
Yomif Kejelcha, who ran 1:59:41 in his marathon debut behind Sawe, is reported to train in the 180–200 km per week range, retaining significant track and half-marathon work alongside his marathon block.

The takeaway: there is no single "correct" elite mileage. There is a range. The athletes who go close to two hours all sit somewhere between 180 and 300 km/week, and the difference between them appears to be more about durability, recovery and economy than about raw volume.

What the training week looks like

The structure varies by coach, but a typical sub-2:01 marathon week includes:

Two key sessions a week. One is usually a long run with marathon-pace work embedded — for example, 35–40km with 25km at marathon pace. The other is a track or hill session at threshold or VO₂max intensity — for example, 10 × 1km at 5K pace with 2 minutes recovery, or 20 × 400m at mile pace.
A full long run every 7–14 days, usually 35–42km. The Kiptum-style "long run at marathon pace" approach is a recent variant; older Kenyan training did long runs at a moderate, even pace closer to marathon-effort minus 20–40 seconds per kilometre.
Daily easy doubles. Most elites run twice a day, six days a week. The morning run is the longer one (typically 14–18km easy), the afternoon shorter (6–10km easy or recovery). The cumulative load is what makes the weekly numbers add up.
One full rest day or near-rest day per week. In some camps that is a Sunday off; in others it is a 30-minute jog plus strides. The willingness to take a real rest day is correlated, in coaches' anecdotal experience, with whether an athlete makes it to the start line healthy.
Strength work twice a week. Single-leg loading, hip-stability work, light plyometrics. Almost no athlete in the modern elite cohort does heavy barbell training in the marathon-specific phase; the trend has shifted toward low-load, high-quality, sport-specific strength.

The thing that does not appear in the training plan, but matters enormously, is the cumulative total over years. The elite sub-2 athletes have all been running serious mileage since their late teens. By the time they line up at a major marathon in their late twenties, they have ten or more years of altitude training in their bones. There is no fast track. The aerobic adaptations that allow this performance are not built in a six-month block; they are built in a decade-long block of which the last six months happen to be the marketed bit.

The role of altitude

The three runners who finished inside the previous world record at London 2026 — Sawe, Kejelcha, Kiplimo — all live and train at significant altitude:

Sawe trains primarily in Kapsabet (~2,000m).
Kejelcha is based in the Ethiopian highlands around Addis Ababa (2,300m+).
Kiplimo trains in the Kapchorwa region of Uganda (~1,800–2,000m).

Altitude has two relevant effects. First, the chronic stimulus of training in lower-oxygen air drives the body to produce more red blood cells, which raises oxygen-carrying capacity (and therefore VO₂max) when the athlete returns to sea level. Second, every easy run at altitude is, metabolically, a slightly harder run than it would be at sea level, so the aerobic adaptations come faster.

The legendary Kenyan and Ethiopian training camps — Iten, Kaptagat, Eldoret, Kapsabet, Bekoji, Sululta — are not magic. They are places that combine altitude, easy access to soft trails, year-round mild weather, and a dense network of training partners. Take any of those four ingredients away and the camps lose much of their effect. Take three of them away and you have a normal city.

Diet, lifestyle and the boring stuff

The diet of elite Kenyan marathoners is, by Western sports-nutrition standards, surprisingly simple. The staples are ugali (a stiff maize-flour porridge), rice and beans, chapati, vegetables (sukuma wiki / collard greens), milky tea, and meat in modest quantities. Macronutrient breakdowns of typical Kenyan elite diets place them at roughly 75–80% carbohydrate, 10–15% protein, 5–10% fat — heavily carbohydrate-loaded, by current global recommendations.

Day-to-day life in the camps is, by design, dull. The schedule is build around training, eating, sleeping and napping. Athletes typically run their first session at 6 a.m., eat, nap, do strength or a second run mid-morning, eat, sleep through the heat of the afternoon, run again at 4 p.m., eat, and sleep early. There are no clubs, no late dinners, no driving around. The contrast with the lives of most amateur runners is not subtle — and it is, in the coaches' view, one of the underrated reasons the model works.

What this means for the rest of us

There are two ways to read all of this. The first is dispiriting: most of the things that make a sub-2 marathon possible — the genetics, the altitude, the developmental years of running on dirt roads, the lifestyle — are not on the menu for someone working a desk job and training around it. That is true.

The second is more useful. The principles under the elite training are the same principles that govern improvement at any level. More mileage than you currently run, conducted at a thoughtful balance of easy and hard intensities. A long run that is genuinely long. Two key sessions a week with real recovery between. Strength and durability work that protects the running. A diet biased toward carbohydrate. Sleep. Patience over the multi-year horizon. None of those things are gated to people with VO₂max numbers in the 70s. They will make a 4:30 marathoner faster as reliably as they made Sawe faster. The numerical ceiling is different for everyone; the gradient up the curve is the same.

If you want to see what 3% of running-economy gain — the realistic effect of modern super-shoe technology — would look like for your marathon time, plug your current PB into the calculator at runpace.co and shave 2–3% off. That is the delta the lab numbers translate into. The rest of the gap to a sub-2 marathon is, fortunately or unfortunately, decades of training and the right grandparents.