Race predictor

How the race predictor works

The predictor uses the Riegel formula, published by engineer and runner Pete Riegel in the 1970s. It models how your time grows as the distance grows:

T1 is your recent race time, D1 its distance, and D2 the distance you want to predict. If distance doubled and speed stayed constant, time would double too. The exponent of 1.06 captures the fact that nobody holds the same speed over twice the distance, so time grows slightly faster than distance. This tool uses 1.06 when predicting a longer race and a slightly gentler 1.055 when predicting a shorter one.

As an example, a 5K in 25:00 predicts a 10K of about 52:07, because the distance doubles and 2^1.06 is about 2.085.

The range

Alongside the single prediction you get an optimistic and a conservative time. The range is about 2.5% either side when the two distances are close, and widens up to 10% as they move further apart, because uncertainty grows with the gap.

Limits

• Distances close together are more reliable. A 10K to half marathon prediction is stronger than a 5K to marathon one.
• Longer predictions run optimistic. Marathon success depends on weekly mileage, long runs, fueling and pacing, none of which a 5K shows. If your training doesn't match the target distance, lean toward the conservative time.
• Use a real race effort. A training run or a race that was cut short gives a weaker answer, and heat, hills and wind in the source race will skew the result.
• Everyone's endurance differs. Naturally fast, shorter-distance runners often fade more than the formula expects.

Once you have a goal, plan your splits or see your training paces.