Running pace is elapsed time divided by distance. That sentence is simple, but reliable results require an explicit distance unit, a whole elapsed time, and rounding only after the complete ratio is known.
Start with two known values
For pace, divide elapsed seconds by distance and scale to one kilometre or mile. For finish time, multiply distance by pace. For distance, divide elapsed time by pace. Keep time in seconds until the last display step so an hour boundary cannot disappear inside separate minute arithmetic.
Normalize distance once
PaceCalc stores accepted input as whole micrometres: one kilometre is 1,000,000,000 µm and one mile is 1,609,344,000 µm. A 5K is therefore exactly 5,000,000,000 µm. Metric and imperial outputs then come from the same canonical distance rather than from one rounded display feeding another.
Round the display, not the formula
A pace can contain a fraction of a second per unit. PaceCalc keeps that fraction as a reduced rational number, calculates speed and every cumulative split from it, and only then rounds a shown clock to the nearest second. This prevents a displayed 5:41 pace from being multiplied back into a noticeably wrong half-marathon finish.