How Running Pace and Race Predictions Work

Quick Answer

What running pace means and how it differs from speed is the first thing to settle: pace is time divided by distance, read as minutes and seconds per kilometre or per mile, while speed is distance over time. This pace calculator, or race pace calculator, takes a distance and a finish time and returns both, plus speed in km/h and mph, plus predicted 5K, 10K, half marathon, and marathon pace and finish times. For example, 5 km in 25 minutes is 5:00 per kilometre, 8:03 per mile, and 12 km/h. The race time prediction uses the Riegel formula, which raises the distance ratio to the power 1.06, so the predicted pace slows as the race gets longer rather than staying flat.

How Pace Is Calculated: Pace Versus Speed

Total your time in seconds, divide by the distance, and you have pace. Speed is the same relationship inverted: distance over time. Runners think in pace because a race target is a time, and holding 5:00 per kilometre is a concrete instruction you can check on a watch every kilometre. Treadmills and cycling computers use speed instead, which is why converting between the two comes up constantly. Choosing between pace per kilometre versus pace per mile is just as arbitrary: a mile is 1.609 kilometres, so mile pace is always the larger number, and the tool prints both.

The Riegel Formula

How a shorter race predicts a longer one was settled by Peter Riegel in 1977: predicted time equals your known time multiplied by the ratio of the distances raised to the power 1.06. The exponent is the whole idea. If running were purely a matter of holding one speed, it would be exactly 1, and doubling the distance would exactly double the time. At 1.06, doubling the distance multiplies time by about 2.08, which matches the few percent of slowdown real runners show over longer races. It is a small number with a large cumulative effect by the time you reach a marathon, which is why the same endurance model serves as a general race time predictor and as a marathon time predictor in particular.

How Far to Trust the Prediction

The model is most reliable within about a factor of three of the distance you ran. A 5K predicts a 10K well and a half marathon acceptably; a marathon is more than eight times the distance and the prediction quietly assumes marathon-specific training you may not have done. Runners without a long-run base routinely finish 20 to 30 minutes behind their predicted marathon. The prediction describes potential from your current fitness, not what would happen if you started one tomorrow.

Examples

5 km in 25:00. For example, that is 5:00 per kilometre, 8:03 per mile, 12 km/h, and predicts a 10K near 51:50.

10 km in 45:00. For example, 4:30 per kilometre predicts a half marathon around 1:39:40 and a marathon around 3:27:40.

Half marathon in 1:45:00. For example, 4:58 per kilometre predicts a marathon near 3:38:50, which is inside the range Riegel handles well.

Common Mistakes