You say the "shovel" method, being faster leads to less strength and connectivity.

As far as not heating the surrounding materials, when soldering RC electronic stuff, I think that is a good thing. So long as the joint itself gets hot enough to flow the solder, I am satisfied with that. I'm only trying to make the joint, not promote global warming.

I gather you believe bad results will show up from quickly soldered joints. How long will it take for those bad results to show? Years? Decades? Centuries? I've got joints on brushed motors 10 years old that still appear fine to me. My most recent Tekin RX-8 install is starting its third season and not showing ill effects now running on 6S LiPo.

I did some searching online and found a research paper that compared cooling rates on solder joints ranging from furnace cooled to quenched.

Here's the link: http://link.springer.com/article/10.1007%2FBF02669524

I didn't spend the $40 to read the whole paper but the abstract mentions that while faster cooling decreases the ultimate strength of the solder it increases the fatigue life. So this research confirms your statement that quickly cooled joints have lower strength than slowly cooled joints. Reading the preview it seems that this research is concerned with stresses induced by differential expansion of PC boards and components mounted there-on. Doesn't sound like that really applies to the RC joints that end users make. Perhaps the body of the article will say how much lower the strength is but knowing that isn't worth $40 to me.

For me, given that RC electronics joints are not under stress (or shouldn't be!) but there is a lot of vibration in an RC car, I think the tradeoff of greater fatigue resistance for lower strength is a good one. YMMV.