On reasoning (aka the Quote of the Day)

[plural]

[a comment to a user in motorcycles by foobiwan]

That people who are likely to do X are also likely to do Y does in no way imply that if you do X you will be more likely to do Y.

Fifty percent of the people who engage in unprotected sex for long periods will get pregnant. However, no matter how much unprotected sex I have, I will not get pregnant.

Comments

[girl-on-a-stick.livejournal.com]

Correlation does not equal causation.
Debate101

[plural.livejournal.com]

indeed

but such a wonderful illustration

[nova-starr.livejournal.com]

this is actually a fallacy of division, not a post hoc.
logic101.

[budhaboy.livejournal.com]

Actually...

The correllation is not causation is better illustrated by this example:

Suppose the incidence of rape in a given month is correlated with the sales of ice cream in the same month.

It would be improper to conclude that ice cream in some way makes people want to rape.

It would not be improper to include, that if you are in a month that experiencing high ice cream sales, you would also be in a month that is likely increasing high incedences of rape.

Applying this to your particular example, it is not unreasonable to conclude that if you do X you will be more likely to do Y...

As for your tounge in cheek example: it's more of the same 'there is no truth because I'm playing with definitions, and your presumptions upon them' you always trot out. If you venn diagram you statement, you'll note that your are working from a supposition that is in the null set (that you can actually get pregnant), a place, you may remember from Philosphy 101 in which ANY conclusion can be reached.