Research · 4 min read · 2026-10-12
Can you predict your marathon from a 10K?
The most common formula works well up to the half marathon, but predicted marathon times at least 10 minutes too fast for half of 2,303 recreational runners.
Many runners use a 10K or a half marathon to work out what they can run in a marathon. The most common formula works well for shorter races, but a study of more than 2,000 recreational runners showed that it is often too optimistic for the marathon.
The formula
In 1981, the engineer Peter Riegel described how race time grows a little faster than distance. His formula multiplies your time by the ratio between the distances raised to the power of 1.06. A runner with a 50-minute 10K gets a predicted half marathon of about 1:50:20 and a marathon of about 3:50.
What the researchers did
Researchers in New York collected race results and training information from 2,303 recreational runners through an online survey. They tested how well the formula predicted each runner's actual times, and built new models from information a recreational runner normally has.
What they found
The formula was accurate for races up to the half marathon. For the marathon, it predicted times that were too fast: for half of the runners, at least 10 minutes too fast. A model based on one earlier race and information about the runner did clearly better, and a model based on two earlier races did better still.
Weekly mileage and interval training were linked to faster times at all distances, while tempo runs were more strongly linked to the shorter races.
What it means for you
A 10K tells you a lot about your 10K and your half marathon. It tells you less about the marathon, because the marathon also tests how well you hold up after two or three hours. If you start a marathon at a pace calculated from a short race alone, there is a real risk of slowing down badly in the second half.
A recent half marathon is a better guide than a 10K. Your long runs and weekly mileage also say a lot about what you can hold for 42 km.
What the study can't tell you
The data came from a survey, so results and training were reported by the runners themselves. The models are built on runners who chose to answer, not a random sample.
How we use it
Targets in PX7 Training come from your own races and runs, and a reference closer to your race distance carries more weight. A marathon pace calculated from a short race is treated as an estimate, and your long runs before the race show whether it holds.
Sources
- Riegel (1981), Athletic records and human endurance. American Scientist.
- Vickers & Vertosick (2016), An empirical study of race times in recreational endurance runners. BMC Sports Science, Medicine and Rehabilitation.
Written by Jørgen Larsen for PX7 Training.