You're assuming that the system is the same in both cases. It's not. 1. You run the A/C all day: the gas stays both cold, and dry. Cold, dry air is, it turns

You're assuming that the system is the same in both cases. It's not. 1. You run the A/C all day: the gas stays both cold, and dry. Cold, dry air is, it turns out, ok as an insulator. Not very good thermal conductivity at all. 2. You turn the A/C on at the end of the day in a warm, humid environment. The air is moist, which causes two compounding problems: first, the volume of air you need to cool just holds more energy than it would if it was dry. But second, because moist air is more thermally conductive, it's better at coupling heat in from the outside, so more energy is coming into the system as you're trying to cool it. At least, that's the argument. I haven't run the numbers to check orders of magnitude or anything. I wasn't expecting the difference in thermal properties of dry and moist air to be that significant, but there are some very interesting numbers at https://www.engineeringtoolbox.com/moist-air-properties-d_1256.html https://www.engineeringtoolbox.com/moist-air-properties-d_12....