"Ergo, a drone should stay the f*ck away from the trajectory of any aircraft that isn't stationary on the ground. There really is no argument, none whatsoever, for endangering those in the aircraft, however slight anyone deems the risk.
The rest is detail."
I think you missed a minor detail. The subject of the research paper was about the effects of flying and falling drones of three specific models, onto people on the ground.
Your other assertion that "if a drone is present, there is a likelihood of damage" is a bit spurious if technically correct. I'd used "possibility" rather than "likelihood" since possibility implies that an accident may happen more most likely won't, while "likelihood" implies far more certainty of an accident occurring. Even assuming that reported accidents are only 10% of the total of actual accidents, compared to the number of drone flights, the odds are pretty small of an accident occurring.