I think you'll find you are controlling the boil rate, not the temperature. If you are reading a higher temperature, it's because superheated steam bubbles are hitting your temperature probe, not because the liquid is hotter. The best placement for a temperature probe is where no bubbles can get to it. Hard to do in boiling liquid. Boiling occurs because the liquid flashes to steam on the bottom of the kettle (or the surface of the immersion heater) and forms bubbles that float to the surface. For that to happen, the mass of liquid must get up to the boiling point so that it can no longer cool the surface of the heating element below boiling point. The hotter the surface of the heat source, the bigger and faster the bubbles. But the liquid temperature will not go above the boiling point determined by the chemistry of the liquid and atmospheric pressure. All it can do at that point is change to gaseous state (steam), which is exactly what is happening at the point of contact with the heat source. If the surface in contact with the liquid gets below that temperature, the bubbles cease to form and boiling stops. A lower boil rate evaporates less liquid than a higher rate because less of the liquid is flashed to steam at the surface of the heat source. Basically, boiling temperature is, or it isn't. No control on it. Because the gas burner also heats the side of the pot, this will help bring the liquid mass up to boiling temperature quicker (more surface area being heated), but the side of the pot rarely achieves boiling temperature or exceeds it, so no steam flash occurs on the side of the pot. But you are correct that a lot of heat goes around the pot and is wasted.
Right or wrong, I have an aversion to immersion heaters, especially with our hard water and the risk of scorching if the element gets too hot in the wort. No different from scorching it in the pot with a propane burner, I know, but a lot of wattage can be put in a small package, and in this case more surface area is the smarter plan. I've learned to be gentle bringing wort up to boil, the hard way. But once I get it going, then I roll it to reduce to target volume as quickly as possible. Encased elements are better than exposed cal-rods, and there are certainly ways to avoid scorching with immersion elements, but I think I prefer a more even heat on a larger surface area that I can control the wattage output from. Not hard to do, really, and relatively inexpensive by simple variac control on the AC power to the heat elements. I feel like I want a hotplate with a diameter close to that of the bottom of my kettle. That's turning out to be hard to find within the price range I was hoping to meet. The only reason I can give is because I want maximum surface area so I can use less heat to do the same thing quicker without scorching. I will still get rid of the wasted heat going around the pot, albeit, that then lends to the sides of the pot being a heat sink and allowing the wort to cool against it. One of those brewing conundrums that is fun to learn from.
