I used to hear about it often from brain teaser quiz books, along with other science trivia, when I was a kid. After a while, I heard it was a common statement, back in medieval and Renaissance days. That kinda makes sense, the only refrigeration then was outside air temps below freezing, and you can really see; hot water vs cold water, stored in wooden (insulating from the sides) buckets, with evaporative cooling outdoors from the hot water's surface could presumably cause it to ice over first. If you browse the wikipedia article's discussion page, and some of the references, you'll see that people are using the Mpemba effect to prove that science can't prove everything empirically, or that western science has an anti-African bias, or an anti-student's observation bias, or some such foo. So the whole issue rapidly degenerates into a big mess.
One explanation for the Mpemba is evaporative cooling from the upper surface, such an effect should be minumized in small freezer where the air becomes moisture saturated. The wikipedia article used to reference a peer-reviewed article where that was tested and found not to be the case, the Mpemba effect remains in small freezers, but it's missing from the article now.
One thing's for sure, if you try to prove or disprove it, you're likely to get yelled at by the peanut gallery.