There are some specialized cases where this can be made to happen. Mostly, it happens because of variables that are hard to control. If the hot water is near boiling, a lot of water evaporates from the hot container, thus there is less water to freeze. If a hot container and a cool container are placed on ice, the ice melts to form a more perfect contact with the hot container, increasing the heat transfer on the hot container.
When all things are equal, the hot water loses it's heat for a period of time until both containers are the same temperature. At that point, they will lose heat at exactly the same rate, until they freeze. The hot container doesn't somehow remember that it used to be hotter and lose it's heat faster than it normally would.
Given that the hot water will take some time to reach the same temperature as the cold sample, that time is lost.
This phenomenon has been argued and tested over since Aristotle's time.
From what I have gathered, experiments can be set up that will seem to verify it, but there is always a "yeah, but" moment attached.
Try it this winter. Set several matching glasses with the exact same amount of water in them with varying temperatures cold, lukewarm, warm, hot, near boiling on your porch this winter and set a timer to remind you to check it once every 5 minutes. See which ones skim over first. Weigh them to ensure each glass is the same and the amount of water before AND after the experiment is the same. You will then be the definitive expert on the subject and you can either bow to my superior knowledge or demand my subservience to your most magnificent being, the keeper of all knowledge of importance.
John
http://www.scientificamerican.com/article.cfm?id=is-it-true-that-hot-water
