Analysis
Statistical significance
A result is statistically significant when it is larger than the noise you would expect by chance. 'Significant' is not a synonym for 'important' — and on a fair lottery, almost everything you can measure is noise.
Noise first
The first question about any extreme statistic is: how often does a random process produce something this extreme? That is exactly what the simulations on this site compute — the share of simulated random histories that match or exceed the observed value.
Correction and consistency
Testing many statistics requires correcting for multiple comparisons, and a real effect should show up across time periods, not in one lucky stretch. The prediction lab applies both checks — and the reality check always reports simulated ranges next to observed values so you can judge the noise yourself.