Tools Race Time Predictor
What could you run?
Enter one recent race and read the equivalent times at 5K, 10K, half marathon and marathon. It uses Riegel’s published formula, shows you the formula, and is clear about the difference between an equivalent performance and one you are trained for.
h:mm:ss or mm:ss — read as 48:30
Equivalent times from your 10K
Riegel baseline, exponent 1.06
| Distance | Riegel baseline | Pace | Focus |
|---|---|---|---|
| 5K | 23:16 | 4:39/km | |
| 10Kyour result | 48:30 | 4:51/km | |
| Half marathon | 1:47:01 | 5:04/km | |
| Marathon | 3:43:07 | 5:17/km |
Your marathon estimate
3:43:07
That is the time equivalent to your 10k, at 5:17/km. Equivalent is not a promise. It assumes you have trained for this distance and that conditions are comparable — the formula knows only your two numbers, not your long runs, the heat, or the hills.
- This is a long extrapolation — more than four times the distance you ran. Equivalence formulas stretch furthest here, and this is where they are least reliable.
The default is Riegel’s published exponent, 1.06. How this works
How this works
The model is Riegel’s: T2 = T1 × (D2 ÷ D1)^1.06. Take your time at one distance, multiply by the ratio of the distances raised to the power 1.06, and you have the equivalent time at another. The exponent 1.06 is Riegel’s published value, fitted across race results, and it is this tool’s default.
Where RunV departs from that, it says so. The optional weekly-mileage buttons move the exponent to 1.09 or 1.03. That direction is defensible — runners on higher mileage tend to hold pace better as distance grows — but those two numbers are our assumption, not published methodology. They are never the default, and the Riegel baseline stays in the table beside them so you can see exactly what the assumption changes.
An earlier version of this page applied a mileage adjustment silently and showed only the adjusted figure. That is the specific thing this version does not do.
A worked example
A 48:30 10K. In seconds that is 2,910. For the marathon, the ratio of distances is 42,195 ÷ 10,000 = 4.2195. Raised to the power 1.06 that is about 4.600, and 2,910 × 4.600 gives roughly 13,387 seconds — 3:43:07.
Ask for the same runner’s 5K and the ratio is 0.5, giving about 23:15. The model always returns your input time at your input distance exactly, which is the easiest way to check it has not drifted.
What this cannot tell you
Equivalent is not achievable. The formula assumes you have trained appropriately for the target distance and that conditions are comparable. A 10K runner who has never run beyond 15 km has a marathon equivalence, not a marathon.
Uncertainty grows with the stretch between distances and with the age of your result. We describe that in words rather than a percentage, because a percentage here would read as a probability and there is nothing behind it that would justify one. There are no confidence intervals on this page, and no chance-of-success figure, because the model does not produce either.
It also has no terms for heat, wind, altitude, hills, injury or how your training is actually going. We have not invented correction factors for them.
Sources
Riegel, P. S. (1981). Athletic Records and Human Endurance. American Scientist 69(3), 285–290.
T2 = T1 × (D2 ÷ D1)^1.06. The exponent 1.06 is Riegel's published value and is this tool's default.
RunV's own choice, labelled on the page
The optional weekly-mileage scenarios (exponent 1.09 / 1.03) are an exploratory assumption, not published methodology, and are never shown without the baseline beside them.
Method version 2.0 · sources and method last reviewed by a person on 2026-09-08.
That review date is not a deploy date. It changes when someone checks the method, and at no other time.
Asked often
- How does the race time predictor work?
- It uses the Riegel formula — T2 = T1 × (D2/D1)^1.06 — which models how pace decays as race distance grows. The exponent 1.06 is Riegel's published value and is what this tool uses by default. You can optionally compare a weekly-mileage assumption that moves the exponent to 1.09 or 1.03, but those two numbers are RunV's own and are labelled as such on the page, with the published baseline always shown beside them.
- How accurate are race time predictions?
- Good for one step up in distance from a recent, honest race effort. A 10K predicts a half well; predicting a marathon from a 5K stretches the model, and endurance-specific training still has to be done. Treat predictions as targets to train toward, not promises.
- Why does weekly mileage change my prediction?
- Two runners with the same 5K time rarely run the same marathon. The one running 70km weeks has the aerobic base to hold pace late; the one on 25km weeks usually fades. RunV's optional mileage scenarios illustrate that idea, but their exponents are our own assumption rather than published or fitted figures, and the Riegel baseline is always shown beside them.
A prediction is a gap. RunV closes it.
The predictor shows what your fitness is worth today — RunV builds the adaptive plan that turns it into a finish line, then see what your goal takes.
