High speed crash on maiden flight--design flaw or bad settings?

Putting a layer of PET/PP on the foam by either applying packing tape or clear document-type laminate makes a huge, huge difference.

Now I am not sure about my travels anymore. Maybe I do have a bit more. Have to look at it.
Anyway; if it behaves as expected nothing is wrong.
Sidenote: I usually run the mixing a bit above 0.5; if you need full pitch you usually do not need full aileron and vice versa. So I find that MIXING_GAIN around 0.7 is fine for my flying style.

Too fast a throttle slew rate and too much throttle during launch induces torque roll. The motor+prop accelerating clockwise also rotates the entire airframe counterclockwise until enough airspeed has built up to make the ailerons effective. In that case a bit slower throttle slew rate helps since it gradually applies more torque when airspeed builds up. The default of 100 is not that bad at all and I would only increae it after veryfying that torque roll is not an issue.

For my settings see the TKOFF_* settings in my param file posted above. THR_MAX is 80%; I usually run that plane on 3S but I still have the option to use 2S 21700 for long duratino flights when needed.
Once again: those are not the gold standard so please evaluate the settings and decide if they fit your purpose.

Just wanted to revisit this thread to say that by October, I had solved all the problems and could launch, fly, and land both of the Nano Goblins I’d built with a 100% success rate. The larger Nano Goblin has a 6400 mAh battery and can fly for 2 hours 10 minutes (total distance: 120 km) while recording footage. I’ve flown both planes dozens of times, and they both work like a charm.