Race Time Predictor
Predict your race times based on recent performance
Race Time Predictor
Predict your race times based on recent performance
What is a Race Time Predictor?
A race time predictor estimates how fast you can run a target distance based on a known recent performance at another distance. If you recently ran a 10K in 50 minutes, the predictor can estimate your likely marathon finish time — saving you from the costly mistake of starting a marathon at an unrealistic pace.
This calculator uses the Riegel formula, developed by Peter Riegel and published in American Scientist (1981). The formula is: T2 = T1 × (D2/D1)^1.06. The exponent 1.06 captures the well-established fact that performance degrades at longer distances — a runner who is 10% faster over 5K than another runner will not be 10% faster over a marathon. The fatigue effect compounds with distance.
The Riegel formula is the most widely used prediction model in running, used by coaches, race organisers, and running apps worldwide. Its primary limitation is that it assumes similar training preparation for both distances. A 5K specialist who has not done long runs will underperform the prediction at marathon distance.
How to Use This Calculator
Typical Race Equivalency (Average Club Runner)
- • 5K in 25:00: → 10K ~52:00 | HM ~1:55 | Marathon ~4:02
- • 10K in 50:00: → HM ~1:51 | Marathon ~3:54
- • 10K in 45:00: → HM ~1:39 | Marathon ~3:30
- • HM in 1:45: → Marathon ~3:41
- • HM in 2:00: → Marathon ~4:13
Why Race Predictions Are a Planning Tool, Not a Promise
Coaches use race predictors mainly to set training paces rather than to promise a finish time — a predicted marathon pace becomes the target for long runs and tempo sessions months before race day, which is a far more valuable use of the number than treating it as a guaranteed outcome.
Over-relying on a prediction without matching training to it is a common way runners get hurt: a 5K specialist who predicts an optimistic marathon time but hasn't built the long-run endurance for it typically "hits the wall" well before the predicted finish, since the formula assumes equivalent preparation across distances that the runner may not actually have.
Individual physiology also shifts how well the fixed 1.06 exponent fits a given runner — someone with a higher proportion of fast-twitch muscle fibres tends to fade more than the formula predicts at longer distances, while a naturally endurance-oriented runner with more slow-twitch fibres may outperform the prediction at marathon distance specifically.
?Frequently Asked Questions
For predictions between similar distances (e.g. 5K to 10K, or 10K to half marathon), the Riegel formula is typically accurate within 2–5% for well-trained runners. Accuracy decreases significantly when predicting across very different distances — e.g. 5K to marathon — because the physiological demands are fundamentally different. A 5K relies heavily on VO2max, while a marathon is primarily a fat metabolism and pacing challenge.
The 1.06 exponent was derived by Peter Riegel from analysis of world record performances across distances, published in American Scientist (1981). It represents the average rate at which human performance degrades with increasing distance. Some researchers have proposed alternative exponents (Cameron 1985 used distance-specific exponents), but 1.06 remains the standard for its simplicity and reasonable accuracy.
A 10K or half marathon time gives more accurate marathon predictions than a 5K, because the physiological systems (aerobic capacity, fat metabolism, muscular endurance) involved in a 10K or half marathon are more similar to marathon demands. Using a 5K time will typically give an optimistic marathon prediction unless you have specifically trained for marathon endurance. A recent half marathon is the gold standard input for marathon prediction.
The 10% rule — not increasing weekly mileage by more than 10% per week — is a widely used guideline to reduce injury risk. It was popularised by running coaches in the 1980s and is referenced in most recreational running programmes. Research from the British Journal of Sports Medicine (2014) supports gradual progression, though the exact percentage is debated. Most coaches recommend 2–3 easy weeks for every 3–4 weeks of progression.
Different running apps and coaches use different prediction models — some use the Riegel formula with the standard 1.06 exponent, others use McMillan's tables, VDOT-based tables (Jack Daniels), or Cameron's distance-specific exponents. Each was built from a slightly different dataset and mathematical approach, so predictions for the same input time can reasonably differ by several minutes at marathon distance. None is definitively "correct" — they're all statistical approximations from different populations.
No. The formula uses only your input time and target distance — it assumes similar course conditions (flat, moderate temperature) for both the reference performance and the predicted race. A hilly or hot target race will realistically run slower than the flat-course prediction suggests, and a downhill or ideal-weather race can run faster. Treat the raw prediction as a flat-course, ideal-conditions baseline and adjust for known course and weather differences.
Treadmill times are generally 1-3% faster than equivalent outdoor road times at the same effort, since there's no wind resistance and the belt provides some assistance to leg turnover. If using a treadmill time as your known input, expect the prediction to be slightly optimistic for an outdoor race — a treadmill time run at 1% incline is often recommended as a closer proxy for true outdoor effort.