Predict your time at other distances from a recent race
Updated October 2026
Pick your most recent all-out race (or a timed solo effort on a flat, measured course) from the last two or three months, and enter the official or watch-measured finish time. The fatigue exponent controls how much you slow as distance grows; leave it at 1.06 unless you know your endurance lags your speed, in which case 1.07 to 1.08 gives a more conservative long-race prediction. The results list a predicted time and pace (per mile and per kilometer) for 1 mile, 5K, 10K, 10 miles, half marathon and marathon.
T2 = T1 × (D2 ÷ D1)1.06. A 25:00 5K is 1,500 seconds. For the marathon, 42.195 ÷ 5 = 8.439, and 8.4391.06 = 9.5911. 1,500 × 9.5911 = 14,387 seconds, which is 3:59:47, or 9:09 per mile. For a half marathon the factor is 4.6002, giving 1:55:00.
| 5K | 10K | Half marathon | Marathon |
|---|---|---|---|
| 20:00 | 41:42 | 1:32:00 | 3:11:49 |
| 22:30 | 46:55 | 1:43:30 | 3:35:48 |
| 25:00 | 52:07 | 1:55:00 | 3:59:47 |
| 27:30 | 57:20 | 2:06:30 | 4:23:45 |
| 30:00 | 1:02:33 | 2:18:00 | 4:47:44 |
| 35:00 | 1:12:58 | 2:41:00 | 5:35:41 |
| Exponent | Half marathon from 25:00 5K | Marathon from 25:00 5K |
|---|---|---|
| 1.04 | 1:51:44 | 3:49:46 |
| 1.06 | 1:55:00 | 3:59:47 |
| 1.08 | 1:58:22 | 4:10:14 |
| 1.10 | 2:01:49 | 4:21:08 |
Each 0.02 added to the exponent costs roughly 10 minutes at the marathon for a 25-minute 5K runner, which is about the gap the 2016 study below found.
Engineer Peter Riegel showed in 1981 (American Scientist) that world-record times for events lasting about 3.5 to 230 minutes follow a power law, time = a × distanceb, because pace slows a little as the race gets longer. For world-class runners he found a fatigue factor of about 1.08 (1.07732 for men’s running records), so doubling the distance multiplies the time by about 21.08, or 2.11. Most running calculators, including this one, default to the slightly more optimistic 1.06 (21.06 = 2.085). A larger exponent predicts more slowdown, which suits runners whose endurance lags their speed.
A 2016 study of 2,303 recreational runners found the Riegel formula well calibrated for races up to the half marathon but too optimistic for the marathon, by 10 minutes or more for half of the runners. In the same study, typical weekly mileage was associated with faster race speeds at every distance, and a model using two prior races predicted marathon times better than the formula (mean squared error 208 versus 381 for Riegel). Races much longer than about 230 minutes, such as ultramarathons, fall outside the range Riegel fit.
Starting a first marathon at the predicted pace. Because the formula runs fast at that distance for most recreational runners, starting slower and finishing faster is safer. Predicting from a short race only. A mile time says more about speed than endurance; a recent 10K or half is a better base for a marathon. Ignoring the course and weather. Heat, humidity, hills and wind all add time that the formula cannot see. Using a race from months ago after a training break or a big block of training. Treating a GPS-short course as the full distance makes the input too fast.
The formula knows only one time and one distance. It cannot see your training volume, fueling, injuries, age-related changes or how well you pace, and it is not a measure of fitness or health. If you are stepping up to a half or full marathon after being mostly inactive, or have known heart, metabolic or kidney disease, the American College of Sports Medicine’s screening guidance is to get medical clearance before starting vigorous training. Stop running and seek care for chest pain or pressure, fainting, unusual breathlessness or a racing, irregular heartbeat, and in hot races watch for confusion or stopping sweating.