Race Time Predictor
Riegel formula · project times across race distances
Pete Riegel's formula projects a finish time at a new distance from a recent result: T2 = T1 × (D2 ÷ D1)1.06, where T1 and D1 are your known time and distance and D2 is the target. Enter one race you've run, and this tool predicts your 5K, 10K, half-marathon and marathon. Predictions are most accurate near the known distance; jumping from a 5K to a marathon tends to be optimistic.
Source: Riegel, 1981 (American Scientist).
General training estimate, not a guarantee of performance.
The formula
T2 = T1 × (D2 / D1)^1.06
T1, D1 = your known time and distance
D2 = the distance you want to predictConvert any predicted time to splits with the running pace calculator. For a full explanation of the model and its limits, read Riegel race-time prediction explained.
Frequently asked questions
What is the Riegel race-time formula?
Pete Riegel's 1981 formula predicts a finish time at a new distance from a known one: T2 = T1 × (D2 ÷ D1)^1.06. T1 and D1 are your known time and distance; D2 is the target distance. The exponent 1.06 captures how pace slows as distance grows.
How accurate is the Riegel predictor?
It is most accurate when the predicted distance is reasonably close to the known one and when you are equally well trained for both. Predicting a marathon from a 5K usually overestimates your marathon ability because endurance and fueling matter more over longer distances. Treat longer predictions as optimistic targets.
Why is the exponent 1.06?
Riegel fitted race data and found that, on average, doubling the distance increases time by a factor of about 2^1.06 ≈ 2.08 rather than exactly 2. The 1.06 exponent reflects the small, predictable slowdown in pace as races get longer. Some calculators use a slightly different exponent for very long or very short events.
Can I use it for any distance?
Yes, for typical road-race distances from about 1500 m to the marathon. It works best between roughly 5K and the half-marathon. For very short sprints or ultra distances the single-exponent model breaks down.