How race time prediction works
Linear projection simply holds your pace constant: if you ran a 10K in 50:00, it predicts a marathon in 3:30:58. That is optimistic for long races and pessimistic for short ones, because nobody keeps 10K pace for 26.2 miles.
The Riegel formula (Peter Riegel, 1981) models that fatigue: T₂ = T₁ × (D₂ ÷ D₁)1.06. The exponent 1.06 means every doubling of distance costs about 4 % in pace. From the same 50:00 10K it predicts a 3:50:00 marathon — a far more realistic target than the linear 3:31.
When it is accurate. Riegel works best when the reference distance is at least a third of the target and you are trained for the longer race. Predicting a marathon from a 5K usually comes out too fast; predicting from a half marathon is reliable. Heat, hills and course profile are not modeled.
Milestone badges flag when a projected time is within a few seconds of a classic barrier (sub-20 5K, sub-4 marathon and so on). Use the pace chart to see how much faster per mile or kilometer you would need to be.