Skip to main content
AI Fit Hub

Benchmarks, in Running

Marathon Statistics: Finishing Times, Demographics & Records

Average times, records, training volume, risk and race-day factors for the marathon. Every figure below links the source it comes from: World Athletics records, RunRepeat's race-result analyses and peer-reviewed sports-medicine studies.

15 figures. Published Mar 12, 2026. Updated Oct 19, 2026. By AI Fit Hub.

Education, not medical advice. The result comes only from the published formula and your inputs. How we check our formulas.

Statistics

The numbers worth quoting

1

The average US marathon finish time in 2024 was 4:24 for men and 4:51 for women (4:28 and 4:54 in 2019)

From about 10.5 million US race results across 6,352 events, roughly 97% of US marathon results. Average times got slightly faster between 2019 and 2024.

2

The men's marathon world record is 1:59:30 (Sabastian Sawe, London, April 2026), the first sub-2-hour time in a record-eligible race

About 2:50 per km (4:34 per mile). It took the record from Kelvin Kiptum's 2:00:35, set in Chicago in October 2023.

3

The women's marathon world record is 2:09:56 (Ruth Chepngetich, Chicago, October 2024), set in a mixed race

The first women's marathon under 2:10. The women-only record is 2:15:41 (Tigst Assefa, London, April 2026). Chepngetich was later banned after a March 2025 test; results before that sample stand.

5

Qualifiers for the 2004 US Olympic Marathon Trials ran 145 km a week (men) and 116 km a week (women), 75% and 68% of it slower than marathon pace

Survey of 93 qualifiers. Men ran 8.7 sessions a week on average and women 7.1; elite women ran more than national-class women (136 vs 111 km a week).

6

Elite marathon times are fastest at about age 27 for men and 29 for women

From the top ten finishers at each age in the 2010 and 2011 New York City Marathons. Times rise on both sides of that age, giving a U-shaped curve.

7

More than two-fifths of marathon runners hit 'the wall', a severe depletion of carbohydrate reserves that limits performance

The model ties it to glycogen stores, running speed and fuelling, so it is not inevitable: carbohydrate during the race and a realistic target pace both delay it.

9

Runners had 30% lower all-cause and 45% lower cardiovascular mortality than non-runners, about 3 years more life expectancy

55,137 adults followed for a mean of 15 years. The benefit was similar across running doses, including small amounts of running each week; this is leisure-time running, not marathon racing.

10

During endurance events, 30-60 g of carbohydrate an hour is an appropriate target, and events longer than 2.5 hours may benefit from up to 90 g an hour

Blends of different carbohydrates can raise how much the gut absorbs at the higher rates.

11

82% of runners met the criteria for stage 1 or 2 acute kidney injury right after a marathon, and 73% showed tubular injury on urine microscopy

22 runners at the 2015 Hartford Marathon. Markers peaked immediately after the race and had fallen a day later; the authors call for larger studies with longer follow-up.

12

Cardiac arrest during US marathons occurred at about 1 per 100,000 participants (1.01)

From 10.9 million runners in US marathons and half-marathons from 2000 to 2010; across both race types the rate was 0.54 per 100,000. Cardiovascular disease accounted for most arrests.

13

Air temperature is the environmental factor that most affects marathon performance; above an optimum, speed falls and withdrawals rise

1.79 million finishes at six major marathons from 2001 to 2010. The optimal temperature depends on performance level; ozone also mattered, possibly through its link with temperature.

14

Carbon-plated prototype racing shoes lowered the energetic cost of running by 4% compared with established marathon shoes

Measured in 18 high-caliber runners at 14-18 km/h, with shoe mass matched; every runner improved. The effect on race time is smaller than the economy gain.

15

In Olympic and World Championship marathons, medallists ran at an even pace from 10 km onward, and athletes who stayed with the same pack slowed least

673 men and 549 women across nine championships. Women slowed less than men and were more likely to run a negative split.

Key takeaways

The men's record went under two hours in 2026: 1:59:30 by Sabastian Sawe.
US average finish times are about 4:24 (men) and 4:51 (women).
Cardiac arrest is rare, about 1 per 100,000 marathon participants.
Carbohydrate during the race (30-90 g an hour) helps delay the wall.
Heat is the race-day factor that costs the most performance.

Methodology

Each statistic is taken from the source it links: World Athletics for records, RunRepeat's race-result analyses for average times and participation, and peer-reviewed studies for the rest. Figures were checked against those sources on 2026-10-19; where a study measured a specific group (elite runners, one race), the context says so.

Run the numbers next

Sources

  1. Cardiac arrest during long-distance running races , New England Journal of Medicine (2012) — Kim et al.
  2. Leisure-Time Running Reduces All-Cause and Cardiovascular Mortality Risk , Journal of the American College of Cardiology (2014) — Lee et al.
  3. A Comparison of the Energetic Cost of Running in Marathon Racing Shoes , Sports Medicine (2018) — Hoogkamer et al.

Keep the topic connected

General fitness estimates, not medical advice. For a medical decision, ask a doctor.