How we calculate

Every number on this site is either a sourced fact or a clearly labelled estimate. When in doubt, we pick the less flattering number.

Percentiles

Finish times are modelled as a log-normal distribution per distance and sex, anchored to published medians and top-10% cut-offs (RunRepeat, State of Running; PacePercentile). These sources lean towards US and European races. Because you enter your personal best but are compared with everyone's results — including their off days — we subtract 3 percentile points.

“1 in N people alive”

We estimate how many living people have ever finished the distance, then multiply by the share of finishers faster than you. For the marathon we use 80 million (the high end of outside estimates); for the half, 75 million (the high end of our own model: about 6 million finishes a year, accumulated over ~25 years, about two finishes per person). World population: 8.3 billion (Worldometer). Higher counts make you look less rare — that's on purpose. The result is rounded down to 100, 150, 200, 300, 500, 700, 1,000 and so on.

Fewer than 1 in 100 people alive have finished either distance, so “99% of people never have” holds even at the high end.

Equivalent times

Cross-distance estimates use Riegel's formula with exponents from 1.06 to 1.10, shown as a range. They never appear on your share card.

Celebrities and history

Celebrity times come from official results, major news outlets or the cited Wikipedia list. We show only each person's best known time. Historical comparisons are made against winners of your own category, or overall. Early races were sometimes shorter than today's distance; we say so when that's the case.

Privacy

No accounts, no cookies. Your result lives in the link. We count visits and shares anonymously to improve the site.