Skip to main content
AI Fit Hub
Menu

Cardio

Race Time Predictor

Race Time predictor: predict finish times across 5K, 10K, half marathon, and marathon from any known race result using Riegel's formula.

Your numbers

Start from an example

Units
Race you ran
Finish time

Type it as 22:30 or 1:40:00.

Predict for

Pick one or more races.

Result

Marathon prediction
3:50:01

5:27 /km average pace

5K
23:59

4:48 /km

10K
50:00

5:00 /km

Half Marathon
1:50:19

5:14 /km

Your race pace
5:00 /km

Predicted finish times

Riegel: T2 = T1 × (D2/D1)^1.06 — longer races run a little slower per kilometre.

5K
4:48 /km
23:59
10K
5:00 /km
50:00
Half Marathon
5:14 /km
1:50:19
Marathon
5:27 /km
3:50:01
0:003:50:01

Education, not medical advice. The result comes only from the published formula and your inputs. How we check our formulas.

What does the Race Time Predictor calculate, and which formula does it use?

The Race Time Predictor converts one known race result into predicted finish times and paces for 1K, 1 mile, 5K, 10K, half marathon, and marathon using Riegel's endurance formula, T2 = T1 × (D2/D1)^1.06. The 1.06 exponent models how pace slows with distance, and each prediction reports how much per-kilometer pace slows versus your baseline race.

Last updated .

How far off can this be?

In 2,303 recreational runners, the Riegel formula this calculator uses was well calibrated for races up to a half marathon but dramatically underestimated marathon time: it was at least 10 minutes too fast for half of the runners.

From the example 10K in 50:00, the calculator predicts 1:50:19 for a half marathon and 3:50:01 for a marathon. Expect the half to be close; for the marathon, half of runners in that study needed 4:00:01 or longer.

The race times were self-reported in an internet survey. The study's own marathon models, which add training data to one or two prior races, cut the prediction error by 40% to 45% against Riegel (mean squared error 228 and 208 against 381).

Source: Vickers & Vertosick 2016, BMC Sports Sci Med Rehabil, “An empirical study of race times in recreational endurance runners” .

How to use it

  1. Select the race you ran (or Other for any distance) and type your finish time as m:ss or h:mm:ss — even 30 seconds of error shifts all predictions. Use an official chip time, not a watch start/stop.
  2. Choose the races to predict. The predictor uses Riegel's formula (T2 = T1 × (D2/D1)^1.06), which assumes you are as well trained for the target race as for the one you ran, so it is most dependable between nearby distances.
  3. Read each prediction with its average pace, per km or per mile with Units set to US. Riegel makes marathon pace about 14% slower per km than 5K pace; if your real marathons run slower than that, the gap is endurance training, not the formula.
  4. Use predicted paces as references for race-specific sessions: marathon pace for marathon-pace work, half marathon pace for tempo-style runs. Easy long runs usually go slower than marathon pace.
  5. Rerun after each race — your fitness evolves and a fresh data point is always more accurate than extrapolating from a 6-month-old result.

Questions people usually ask

Can this predict my half marathon time?

Yes. Half marathon (21.0975 km) is one of the four target distances the predictor outputs alongside 5K, 10K, and the marathon. Enter any one recent race result and it projects your half marathon finish using Riegel's T2 = T1 × (D2/D1)^1.06. Predictions are most dependable between nearby distances, so a 10K result gives a more reliable half marathon estimate than a 5K.

What formula is used?

Riegel's endurance model, published in American Scientist in 1981: T2 = T1 × (D2/D1)^1.06. The 1.06 exponent captures the non-linear fatigue increase with longer distances — you slow down proportionally more as distance increases.

How accurate are these predictions?

Closest for well-trained runners predicting between nearby distances. Accuracy drops for first-time marathoners extrapolating from a 5K (recreational runners often run the marathon slower than Riegel predicts), and for athletes who are much stronger at one distance than another.

Why does my predicted marathon time seem too slow?

Riegel's formula assumes consistent aerobic fitness across all distances. If you are undertrained for the marathon (short long runs, not enough weekly volume), your actual time will be slower than predicted. The formula models potential, not current race readiness.

Can I use a training run instead of a race?

A hard, evenly paced time trial works. An ordinary training run does not: it was not a maximal effort, so every prediction comes out too slow. Race results give the most accurate predictions because they represent your best effort over the distance.

Is the tool free and private?

Yes. All calculations are client-side. No data leaves your browser.

  • Nutrition TDEE Calculator

    Estimate your daily energy expenditure with Mifflin-St Jeor + activity factors.

  • Nutrition Macro Calculator

    Convert calorie targets into protein, carbs, and fat grams for your goal.

  • Planning Calorie Deficit Calculator

    Estimate required daily calorie deficit for a target timeline and bodyweight change.

  • Planning Body Recomposition Planner

    Plan body-fat reduction pace, deficit targets, and protein needs around a timeline.

All articles

Use it from an agent or code

Any agent can run this calculator by URL: put the inputs in the query string and get JSON back, with the inputs that used defaults and a line to cite (add format=md for markdown). MCP clients reach it through https://aifithub.io/mcp.

https://aifithub.io/api/run/race-time-predictor?known_distance_km=10&known_time_minutes=50

In code, import the same module; it runs where you import it, with no request or key.

import { compute } from "https://aifithub.io/engines/race-time-predictor.js";

Input and output contract, TypeScript types and the guide for agents.