How to use the Race Time Predictor
- Pick the distance you have a genuinely recent, hard-effort result for.
- Enter that time in minutes and seconds.
- Read the predicted times for every other distance, each with its pace per kilometre.
- Trust predictions for distances close to your input more than distant extrapolations.
- Re-check the marathon prediction against your longest recent run before trusting it.
How the calculation works
Riegel's formula states T₂ = T₁ × (D₂ ÷ D₁)^1.06. The exponent above 1 encodes the fact that pace decays as distance grows: doubling the distance takes slightly more than twice the time. Riegel derived 1.06 from thousands of race results across running, swimming and cycling in the 1970s, and it has held up remarkably well as a population average.
The exponent is not universal. Runners with high endurance and heavy weekly mileage behave closer to 1.05, while speed-oriented runners with low mileage sit nearer 1.08. Across the 5K-to-marathon jump that difference is worth several minutes, which is why two runners with identical 5K times can finish a marathon twenty minutes apart. Comparing your actual results at two distances lets you solve for your own exponent.
The marathon is where predictions break down most. Riegel's model assumes the same physiological limiter applies at every distance, but the marathon adds glycogen depletion, thermoregulation and mechanical fatigue that a 10K never touches. A runner who has not built the long-run volume will typically run several minutes slower than predicted. Treat the marathon figure as a ceiling that training must earn.
T₂ = T₁ × (D₂ ÷ D₁)^1.06Source: Riegel PS, 'Athletic Records and Human Endurance', American Scientist 69 (1981); Vickers AJ & Vertosick EA, BMC Sports Science (2016) on individualised exponents.
Worked example
A 10 km time of 50:00 projected to the half marathon.
- T₁ = 3,000 seconds, D₁ = 10 km, D₂ = 21.0975 km.
- Ratio = 2.10975; raised to 1.06 = 2.1955.
- T₂ = 3,000 × 2.1955 = 6,587 seconds.
About 1:49:47 for the half marathon, a pace of 5:12 per kilometre.
Frequently asked questions
Why is my marathon slower than predicted?+
The model assumes marathon-specific endurance you may not have built. Fuelling, heat and long-run volume all cost time it does not model.
Can I predict a 5K from a marathon?+
Yes, and it tends to be more reliable in that direction — short-distance predictions from long results are usually conservative.
What exponent should I use?+
1.06 is the population default. High-mileage runners can use 1.05, speed-focused runners 1.07–1.08.
Does the prediction assume flat courses?+
Yes. Hills, altitude and heat all add time the formula does not account for.
Last reviewed September 1, 2026. We review this page whenever the underlying formula, tax year, published rate or standard changes.