I can't remember where I stumbled upon this but apparently this problem (which my mom had me and my brother experiment in 1995 or 1996) used to be much less popular. Keep in mind, this article was written in 1990.

Still fascinating from a vast number of angles.

it predates the Internet by a fair sight. I remember reading it in real time when it came out. I remember the controversy. I remember all the letters. And yes, it's the "Monty Hall Problem" but at the time, you would have had to know it was called the Monty Hall Problem and you would have had to have been able to guess where, exactly, within the Library of Congress classification system "Monty Hall Problem" would show up.

You also have to consider the crablouse in its native environment. We're talking Parade Magazine here, that little flyer with the easy crosswords, softball interviews and recipes involving Miracle Whip that you get if your hometown paper weren't willing to spring for USA Weekend. There's nothing hard, there's nothing thought provoking and there's nothing controversial in Parade... except the Monty Hall Problem, and the rainstorm problem, which MvS attacked a couple years later and which was not resolved as succinctly in her favor.

Finally, here's how MvS decided to explain one of the stickiest and least-intuitive aspects of statistics to the unsophisticated audience of Parade:

Yes; you should switch. The first door has a 1/3 chance of winning, but the second door has a 2/3 chance. Here’s a good way to visualize what happened. Suppose there are a million doors, and you pick door #1. Then the host, who knows what’s behind the doors and will always avoid the one with the prize, opens them all except door #777,777. You’d switch to that door pretty fast, wouldn’t you?

Does that clarify things? or make them muddier? Or muddier and condescending?

MvS could have blown the entire column explaining why statistics are tricky and how obvious answers are sometimes wrong. She didn't. She snarked. And as a result, she spent the next six months putting out brushfires and trickling the answer out at a pace that made it so nobody remembered the original beef or why it was so controversial.

People hold this event up as an example of why people are stupid, rather than an example of why statistics are complex, why our intuitive understanding of statistics is often wrong and why a better understanding of the underlying problems often requires more effort than people are willing to put into it. Had MvS written "we never found the WMD" in parade in 2004, the responses would have been just as heated, just as self-assured, and just as wrong.