Comfy, this light has a lot of surface area to dissipate heat. The head is huge. The body is fat. The thin pill is not as bad as it seems since it has such a large diameter as well. The surface area of the pill at it's circumference isn't too shabby. So, I am not surprised that you have not noticed "a lot" of sag due to heat. Though, I guess my question is what do you consider a lot? >10%, >20% >30%? When I designed the Bomber, one of my goals was to minimize sag due to heat. I was very happy when it turned out to be <10% after running for 5 minutes at about 80 watts.
A couple of things that you wrote bother me. First, I'm all about the laws of physics. I'm not sure why you think otherwise. There is nothing about using copper heat sinks that have greater mass, thermal conductivity and surface area than the ones they replace that defy the laws of physics. I wrote a fairly detailed post about this some time ago. I'll try to find it and link it. Second, you are right in that sometimes what we do isn't necessary. Most engineering is the science of designing things that work good enough for their purpose plus some safety factor. However, I liken what we do here to hotrodding. Striving to get that extra few percent of output at the margin is freakin cool. For that matter, putting 7 emitters into a flashlight isn't necessary. Most modding that's done here isn't necessary, yet you seem to have a distaste for my particular type of unnecessary.
Anyway, "Good Enough" is relative to each person here. 7 amps through a XM-L2 on the stock pill may work ok for some. Start putting more current to the emitter on that pill and diminishing returns start happening at some point. More amps will not yield more light. Just look at the aluminum star vs Sinkpad copper star graphs. Same goes for pills. Amp for amp, you get more light with copper. You get more light with more area of the pill in contact with the outside of the light too.