The following scenario:
I throw a die 10 times:
6 6 6 6 6 6 6 6 6 6 6
What is the probability of this happening ? (1/6)^10
And despite being much smaller than 0.95 it did happen
What is the likelyhood that the die is 'loaded' ?
Now that is a hypothesis which can be addressed
H0: The observation was due to chance
H1: The observation was not due to chance
Now one can use statistics to determine the likelyhood of H0 versus H1 and often 0.90, 0.95 or 0.99 are used as limits on 'chance' playing a role here.
But even at a 0.95 limit there is a 1 in 20 chance that the observation was due to chance.