That is, just barely at all (a marginal decrease in risk of .22 percent entails that you will on average produce one fewer children every 454 years).
I'm not sure I understand; If you have two dice, each one with 50 sides, and each one representing a form of contraception -- the first being the pill, and the second pulling-out -- and then you roll each one, needing to hit the number 1 on the first and either a 1, 2, 3, 4, 5, or 6 on the second in order for pregnancy to occur... isn't that far more unlikely to occur than merely rolling just one die and hitting the number 1? This is how I think of things sometimes... I'm sure that there's a really simple mathematical formula one could use (and you probably used) to come up with the actual numbers, but, y'know, math has never been my strong suit so forgive me.
EDIT: OK I think I understand now, and it would appear that you're correct. By using condoms or the pill in conjunction with pulling out, you're not much better off than had you simply taken the pill or used a condom. I would have thought that the chance of pregnancy occurring would be further reduced quite drastically by using two forms of contraception, but I guess when one form is already incredibly efficient, well, you can't really do too much better.
I would find an annual risk of accidental pregnancy of 27 percent (no one could be rigorous in technique over the course of a whole year) intolerable unless I sorta wanted kids.
Right. Which is why I very simply said that pulling out was more effective than one might think, albeit not nearly as effective as the pill or condoms. I wasn't implying that pulling out would be something I would consider using as a primary means of birth control. It's simply more effective than one might think. Is that not a fair statement?
But maybe I'm wrong on that point, too; maybe your estimation of its effectiveness, before learning its actual effectiveness, was right on the money. And maybe it's
not more effective than one might originally think, and maybe it's only me. But I would have thought that pulling out's effectiveness would be something more like 30 or 40%, not ~85% ...