While it is true that the thermal mass of the carboy/bucket with liquid is much greater than the surrounding air and that their temperature should dominate, this is true only if you have good thermal contact between the probe and the carboy/bucket. Unfortunately, unless you're using some sort of thermal grease or other goop between the probe head and the carboy, you don't have good thermal contact. In my case, which is using the thermal sensor that came with my temperature controller, I have to try to get a hard cylinder (temp sensor) to make good contact with a curved hard surface (carboy). Even in an ideal setting, that is a very tiny surface contact area, so thermal transfer needs to occur using the air trapped under the tape and/or through the tape itself.
BrewQwest's observations make sense to me. From a thermal engineering standpoint, you'd never attach a temperature probe to a surface unless you used some sort of thermal transfer material, like grease, along with it. If you really want to know and control the temperature of the liquid in an absolute sense, you really need to use a thermowell or attach the probe better thermally. I don't have a good feel for whether putting bubble wrap over the probe before taping it is better or not. To me, making sure air cannot exchange between the ambient and below the tape is more important, and using bubble wrap in my opinion makes that much easier to achieve.