How long does it take to run a marathon?
How long does it take to run a marathon? The answer depends entirely on who you ask. The chart below uses a violin plot to show the full shape of the finishing time distribution — not just an average, but the entire spread from elite runners to first-timers.
What the graph shows
The violin plot displays the distribution of marathon finishing times across a simulated population of recreational runners, modelled after real-world data. Unlike a simple average or bar chart, a violin plot reveals the full shape of the distribution — where runners cluster, how spread out they are, and how symmetric or skewed the pattern is.
The y-axis runs from 2:00 (top) to 7:00 (bottom), representing finishing time in hours:minutes. The width of the violin at any given time reflects how many runners finish around that time: wider means more runners, narrower means fewer.
Reading the chart
The violin shape
The widest point of the violin sits between roughly 4:00 and 4:45, showing this is where the majority of recreational marathon runners finish. The shape is right-skewed: there is a long tail stretching toward slower times (5:00–7:00), while the fast end tapers sharply above the 3-hour mark.
The box plot (inner overlay)
Embedded inside the violin is a standard box plot:
- Median line — the time by which exactly half of runners have finished
- Box (IQR) — the interquartile range, spanning the middle 50% of finishers (Q1 to Q3)
- Whiskers — extend to the furthest values within 1.5× the IQR from the box edges
- Min / Max labels — the fastest and slowest times in the dataset
The 3-hour line
The dashed amber line marks the sub-3 hour threshold — widely regarded as the benchmark of an elite recreational marathon performance. The chart is split into two colour regions:
- Green (top section) — sub-3hr runners, a small fraction of the total field (~2–5% in real races)
- Purple (main body) — the vast majority of runners who finish in 3 hours or more
The percentage labels on the right confirm the split for this simulated dataset.
About the data model
The distribution shown is simulated, not drawn from a single real race dataset. It uses a Gaussian mixture model — five overlapping normal distributions weighted to reflect the shape of the real-world marathon finishing time distribution documented in academic research:
| Group | Weight | Mean time | Std dev | Represents |
|---|---|---|---|---|
| Elite | 3% | ~2:45 | 10 min | Competitive club runners |
| Fast | 12% | ~3:30 | 15 min | Strong recreational runners |
| Core | 30% | ~4:05 | 18 min | Typical recreational peak |
| Mid | 30% | ~4:35 | 20 min | Mid-pack runners |
| Back | 25% | ~5:30 | 35 min | Slower / first-time runners |
KDE (Kernel Density Estimation) with a bandwidth of 15 minutes is then applied to smooth the 4,000 sample points into the continuous violin curve.
Sources
Allen, E. J., Dechow, P. M., Pope, D. G., & Wu, G. (2017)
Reference-Dependent Preferences: Evidence from Marathon Runners.
Management Science, 63(6), 1657–1672.
The foundational study analyzing 9,789,093 marathon finishing times, documenting the right-skewed distribution and the well-known bunching effect just before round-number time goals (3:00, 3:30, 4:00, etc.).
RunRepeat / Jens Jakob Andersen (2019)
Marathon Statistics — Worldwide Average Finishing Times.
Large-scale study of 19.6 million marathon results from 2008–2018, covering 32,335 races across 39 countries. Reports a global average finishing time of 4:29:53.
runrepeat.com
Marathon Handbook (2024)
Average Marathon Time by Age and Sex.
Reports a current overall global average of 4:32:49, with men averaging 4:10:10 and women approximately 4:55. Notes the historical slowdown as participation has broadened.
marathonhandbook.com
RunnerClick / Global Running Statistics (2019)
Marathon Statistics Worldwide.
Identifies the fastest age group overall as 40–49 (avg 4:22:03), and the fastest women's group as 20–29 (avg 4:42:10).
runnerclick.com
Skinnerstad, T., et al. (2018)
Effects of age on marathon finishing time among male amateur runners in Stockholm Marathon 1979–2014.
PMC / European Journal of Sport Science.
Documents that average finishing time has increased approximately 1 minute per year over the history of the Stockholm Marathon, reflecting the democratisation of the sport.
pmc.ncbi.nlm.nih.gov
Comentários
Enviar um comentário