Riegel's formula, T2 = T1 × (D2 ÷ D1)^b, is behind most race time predictors, including the running pace calculator. How well does it do at the very top? This page feeds it the 10,000 m (and 5,000 m) world records and compares the predictions with the actual records.
The comparison
Records as of 27 Oct 2026 (World Athletics list of ratified world records).
| Prediction | World record | Riegel 1.06 | vs record | Riegel 1.08 | vs record |
|---|---|---|---|---|---|
| Men: Half marathon | 57:20 | 57:46 | +0.8% | 58:38 | +2.3% |
| Men: Marathon | 1:59:30 | 2:00:27 | +0.8% | 2:03:58 | +3.7% |
| Men: 10,000 m from the 5,000 m | 26:11.00 | 26:15 | +0.2% | 26:37 | +1.6% |
| Women: Half marathon | 1:02:52 | 1:03:46 | +1.4% | 1:04:44 | +3.0% |
| Women: Marathon | 2:09:56 | 2:12:57 | +2.3% | 2:16:50 | +5.3% |
| Women: 10,000 m from the 5,000 m | 28:54.14 | 29:07 | +0.8% | 29:32 | +2.2% |
A plus sign means the prediction is slower than the record. The women's half marathon and marathon records shown were set in mixed races; the women-only records are 1:05:16 and 2:15:41. A men's half marathon of 56:51 (23 August 2026) was listed as awaiting ratification and is not used.
What the table shows
- With the exponent 1.06, the formula is close for men. From the 26:11.00 men's 10,000 m, it predicts a 57:46 half marathon (0.8% slower than the 57:20 record) and a 2:00:27 marathon (0.8% slower than the 1:59:30 record).
- For women it is further off at the marathon: 2:12:57 predicted against 2:09:56, 2.3% slower. That means the women's marathon record is relatively faster than the 10,000 m record, by Riegel's yardstick.
- The exponent 1.08 is slower in every row. Riegel fitted about 1.08 on world records in 1981 (1.077 for men, 1.083 for women). Today's records, from the 10,000 m upward, sit closer to the 1.06 line.
What this means for everyday runners
World records are set by athletes who train specifically for each distance, which is what the formula assumes. Recreational runners are different. Vickers and Vertosick (2016) studied race times of recreational runners and found the Riegel formula worked well up to the half marathon but gave marathon times at least 10 minutes too fast for half of the runners. So use long-range predictions as optimistic, especially for a first marathon.
The exponent also matters more the further you predict. From a 50:00 10K, 1.06 gives a 3:50:01 marathon; 1.08 gives 3:56:44.