Example: I'd guess that when hormonal birth control fails despite being used perfectly, it could well be related to the person's particular biology. So if it failed for them one year, there'd be a high chance of it failing the next. And if it worked for the previous year, there's a lower chance of whatever goes wrong going wrong.
(I'd be curious to hear from someone who knows what they're talking about, though!)
>How the numbers were calculated: The probability that a woman doesn't get pregnant at all over a given period of time is equal to the success rate of her contraceptive method, raised to the power of the number of years she uses that method.
This does assume that the chance of getting pregnant after a year is independent of the person/people using it, though... presumably there are reasons other than a random roll of the dice why various birth control methods fail, which could well affect the distribution.
(It also seems absolutely crazy to give these statistics "per year" when I'd feel pretty confident in saying that it depends on how often you have sex!)
(I'd be curious to hear from someone who knows what they're talking about, though!)