I found STEPHEN J. DUBNER's The Probability That a Real-Estate Agent Is Cheating You (and Other Riddles of Modern Life) on Steven Levitt interesting, particularly what DUBNER calls this "famous" statement:

If you own a gun and have a swimming pool in the yard, the swimming pool is almost 100 times more likely to kill a child than the gun is.

Not because I'm particularly in favor of gun ownership, but because I'm interested in risk. So I Google "Steven Levitt" gun pool, and I find this big controversy involving John Lott. I haven't been following him or his research very closely, but together with the Mary Rosh business, he's beginning to look pretty kooky. Of course, it's not as bad as the Bellesiles mess, but it's funny that these two big "guns" are to one degree or another such frauds.

Update
In The Bellesiles of the Right? Another firearms scholar whose dog ate his data, Timothy Noah presents some further evidence against Rosh--I mean Lott. (link via Instapundit, who's highly dismissive, of course)