Race Time Predictor: Predict Your 5K, 10K, Half Marathon and Marathon
Enter a recent race distance and time. You will see predicted times for a mile up to the marathon, with the pace per kilometre and per mile for each, using Peter Riegel’s fatigue formula.
Last reviewed
Predicted marathon
3:59:47
5:41 per km, 9:09 per mile
| Distance | Time | Per km | Per mile |
|---|---|---|---|
| 1 mile | 7:31 | 4:40 | 7:31 |
| 5 kmyour result | 25:00 | 5:00 | 8:03 |
| 10 km | 52:07 | 5:13 | 8:23 |
| 15 km | 1:20:07 | 5:20 | 8:36 |
| 10 miles | 1:26:19 | 5:22 | 8:38 |
| Half marathon | 1:55:00 | 5:27 | 8:46 |
| Marathon | 3:59:47 * | 5:41 | 9:09 |
* Riegel’s formula was built for efforts of about 3.5 to 230 minutes. Times outside that range, or predicted from a result outside it, are rough guides only.
A prediction assumes you train for the longer distance as well as you did for the shorter one. Heat, hills and fuelling can move a real race by minutes.
How a race time is predicted
If you can run 5 km in 25 minutes, how fast could you run a half marathon? The most widely used answer is a formula published by Peter Riegel in American Scientist in 1981. He noticed that when you double the distance, the time more than doubles, because you slow down as fatigue builds. He fitted the slow-down with a power law.
T2 = T1 × (D2 ÷ D1)^1.06
Here T1 is the time you ran over distance D1, and T2 is the predicted time over distance D2. The exponent 1.06 is the fatigue factor: with an exponent of exactly 1 you would hold the same pace at every distance, and the higher the exponent the more you slow down.
The limits of the formula
Riegel’s article limited the model to efforts in the “endurance range”, which lasted from about 3.5 to 230 minutes. That covers a mile to a fast marathon, but not an ultramarathon, and it makes a slow marathon (over 3 hours 50 minutes) a rough guide. The calculator marks any time outside that range with an asterisk.
Some runners and websites change the exponent, because 1.06 can give predictions for long distances that are too fast for runners who have not done long-run training. Because of this, the calculator lets you set the exponent yourself.
A prediction also assumes you have trained for the longer distance as well as you did for the shorter one. A 5 km runner who has never run beyond 10 km may not be able to run the predicted marathon.
Worked example
You ran 5 km in 25:00, which is 1,500 seconds. For a half marathon of 21,097.5 metres the distance ratio is 21,097.5 ÷ 5,000 = 4.2195. Raise it to the power 1.06, which gives about 4.6, and multiply by 1,500: the predicted time is 6,900 seconds, or 1:55:00.
For the marathon the ratio is 42,195 ÷ 5,000 = 8.439, and the predicted time is 14,387 seconds, 3:59:47, a pace of 5:41 per km.
Predictions from a 25:00 5 km
| Distance | Time | Per km | Per mile |
|---|---|---|---|
| 1 mile | 7:31 | 4:40 | 7:31 |
| 5 km | 25:00 | 5:00 | 8:03 |
| 10 km | 52:07 | 5:13 | 8:23 |
| 15 km | 1:20:07 | 5:20 | 8:36 |
| 10 miles | 1:26:19 | 5:22 | 8:38 |
| Half marathon | 1:55:00 | 5:27 | 8:46 |
| Marathon | 3:59:47 | 5:41 | 9:09 |
A faster runner shows the same pattern. A 20:00 5 km gives 41:42 for 10 km, 1:32:00 for the half marathon and 3:11:49 for the marathon.
How much the exponent matters. With an exponent of 1.15 instead of 1.06 the same 25:00 5 km predicts a marathon of 4:50:31, nearly 51 minutes slower. That is why a marathon prediction from a 5 km time is the least reliable number on the page. A recent half marathon or 10 km race predicts a marathon far better. A study of more than 2,000 recreational runners found the formula well calibrated up to the half marathon, but it gave marathon times at least 10 minutes too fast for about half of the runners.
Getting a better prediction
- Use a recent, hard, flat race. A time trial in good conditions beats a training run.
- Predict from a nearby distance. A 10 km result predicts a half marathon better than a mile does.
- Compare several results. If your 5 km and 10 km predict different half marathons, trust the one closer to the target distance.
- Allow for conditions. Heat, hills, wind and a poor night’s sleep each add minutes to a long race.
- Set pacing from the pace columns. Use the pace per km or per mile to plan your splits, with the pace calculator.
Related tools: the pace calculator, the target heart rate calculator, the calories burned calculator and the steps to km calculator.
Frequently asked questions
What is Riegel’s formula?
Peter Riegel’s formula predicts a time over a new distance from a time you have run: T2 equals T1 multiplied by the distance ratio D2 over D1, raised to the power 1.06. The exponent builds in the slow-down that fatigue causes as the distance grows.
How accurate is a race time prediction?
It is a guide, not a promise. It works best when the new distance is close to the one you ran, you have trained for it, and conditions are similar. Predicting a marathon from a 5 km time is the least reliable, because endurance training matters more.
What is the 3.5 to 230 minute range?
Riegel’s article limited the model to efforts lasting about 3.5 to 230 minutes. Predicted or entered times outside that range, such as a slow marathon or an ultramarathon, are marked with an asterisk because the formula was not built for them.
Why can I change the exponent?
Some runners and websites use a larger exponent than 1.06 because 1.06 can predict long distances that are faster than less experienced runners can hold. A higher exponent slows the long predictions. The default is Riegel’s own 1.06.
Which result should I use as my starting point?
Use a recent, flat, all-out race or time trial at a distance close to your target. A 10 km race predicts a half marathon better than a mile does. If you have several results, compare the predictions they give for the same distance.
Does it work for walking or cycling?
Riegel’s article covered running, swimming and walking within the same time range, but this calculator is set up for running distances and shows pace per kilometre and per mile. Cycling is not covered by the formula.
Related calculators
- Pace CalculatorEnter any two of distance, time and pace to find the third.
- Target Heart Rate CalculatorEnter your age.
- Calories Burned CalculatorChoose an activity, how long you did it and your weight.
- Steps to Km CalculatorEnter your steps.
- TDEE CalculatorEnter your sex, age, height and weight and choose how active you are.
- BMI CalculatorEnter your height and weight.
Sources
- Peter S. Riegel, Athletic Records and Human Endurance, American Scientist 69(3), May-June 1981, pp. 285-290 (JSTOR record. A full-text search of the article on JSTOR returns the "endurance range" sentence about activities from 3.5 to 230 minutes and the log-log graph of world records, and matches the figures 1.06 and 1.08 in the text. The full text needs a JSTOR account, so the exponent is also cited from the peer-reviewed papers below)
- Peter Riegel (Wikipedia): the formula T2 = T1 × (D2/D1)^1.06 from his 1981 American Scientist article, and its 3.5 to 230 minute range
- Mulligan, Adam and Emig: A Minimal Power Model for Human Running Performance (cites Riegel, P. S., Athletic Records and Human Endurance, American Scientist 69(3), 1981, and gives his exponent as close to 1.06)
- Vickers and Vertosick: An empirical study of race times in recreational endurance runners, BMC Sports Science, Medicine and Rehabilitation 2016 (the formula is derived from Riegel's world-record paper; Riegel cited an exponent of 1.08 for elite runners and 1.05 to 1.06 for male recreational runners aged 40 to 70; the formula was well calibrated to the half marathon but too fast for the marathon)
- Frontiers in Sports and Active Living: Under the Hood: Skeletal Muscle Determinants of Endurance Performance, 2021 (endurance competitions typically run from about 3 to 230 minutes, citing Riegel 1981)
This tool is for information only and is not medical advice. Speak to a doctor or a qualified professional about your own health.