The subject of hot water freezing faster than cold water is often
referred to as
the "mpemba" effect. There was a long series of posts on Phys-L a couple
of years ago. There are many factors involved: amount of dissolved
gases, time at 0 degrees before freezing, amount of supercooling before
actual freezing, etc. etc.
I have a file of all (most) of the postings of the activity from a couple
of years ago. If anyone would care to contact me personnaly, I would be
glad to send it out. I don't want to bother the entire network with the
posting of the file.