If there was a design, it's not clear what the intent was. It seems about twice as precise as it needs to be (i certainly can't perceive 1F°—for all intents and purposes, 70 feels about the same as 69 and 71) and doesn't seem to correlate to any scale that is immediately based off the needs of humans. At least compared to celcius.
He says in his original paper that the top point of his reference scale is 96, not 100 for the point where "Alcohol expands up to this point when it is held in the mouth or under the armpit of a living man in good health". He originally based his scale on 12, and then got more precise by increasing each division by two several times, ending up with 96.
0F .. 100F is about the range of temperatures a human living on earth could reasonably expect to experience without deliberate adventuring. It's not a precise range - plenty of people live in Doha (way above 100F) and in Alberta (way below 0F) - but it's a pretty reasonable approximation.
My comment ends with a note that "it's a reasonable approximation".
The percentage of global population where the 0F..100F range is not a reasonable approximation of the temperature range they will experience is small. It's not perfect - no such range could, when humans live almost everywhere on the planet. But it's not bad ...
I live in the US and can't change my thermostat, so I don't think that's it.
I'm sure I could feel the difference if i split myself between two rooms with one degree difference. I just don't think this is a useful granularity—I typically move the thermostat by 2-5 degrees at a time.
If there was a design, it's not clear what the intent was. It seems about twice as precise as it needs to be (i certainly can't perceive 1F°—for all intents and purposes, 70 feels about the same as 69 and 71) and doesn't seem to correlate to any scale that is immediately based off the needs of humans. At least compared to celcius.