Right. You learn two numbers for the phase changes of water, and we do as well. It's easy to remember two numbers and understand when you've crossed a boundary. Sure, learning 0 and 100 might be easier than 32 and 212, but I don't think that understanding whether a number is smaller or larger than 32 is really harder than understanding if it's smaller or larger than 0. Both are pretty much instantaneous recognitions for a numerically literate person.

My point was merely that the Fahrenheit defines these two points in such a way that the ambient temperatures that we experience generally fall nicely within the range of 0 to 100, and I don't think that this fact is any less compelling an argument than having nicer numbers for the boundaries of liquid water. I'm not saying that Celsius is bad. I'm just saying that the range of liquid water is not a convincing enough argument for me.

For other units of measures, the ease of converting units in metric is a clear win over imperial (or US customary). For temperature, there are benefits to both scales, and neither has as compelling an argument as we see in the meter vs the yard or the kilogram vs the pound. The only really convincing argument for me is that the rest of the world uses Celsius, and I think that is a good enough argument.