I know technology has moved on but I used to get great satisfaction out of being able to complete a plus/minus 10 second NDB hold and approach
The thing is that you could point to any number of 'hey look how skilled I am at this' things but unless they are relevant to
practical flying, they are just a waste of the pilot's time and brain space, and can be regarded as the province of the (many) elitist types which still inhabit the aviation world - particularly the regulatory bits of it.
The protected area of a hold is massive - big enough for a 747 at something like 200kt. Consequently there is exactly zero point in flying one to the accuracy you refer to. It may be fun but a pilot doing that - no matter how good he is - isn't going to have spare brain capacity to keep tabs on everything else he needs to do. Let's say that on hold #7 the surface wind changes resulting in a runway change and he has to dig out a bunch of the 'other' approach plates, while flying this wonderfully perfect NDB hold. The whole job is going to come to pieces - because he had no spare brain capacity. Better still, his fuel goes below reserves while in the hold and then he has to do something else.
The 'art' of flying safely is to manage the cockpit workload so it is minimised, and then you can handle whatever ATC or weather throw at you.
Funnily enough I quite like the ADF - when I used to fly VFR abroad, it made a handy backup for the GPS because very often there were no VORs on the route (the VORs were in CAS). I have a dual-needle RMI displaying the ADF bearing and flying NDB-anything is a piece of cake. But if you look at the cost of a modern reliable ADF (and there is only the KR87 - the rest is nearly all unreliable junk) it's a good few grand and this is not justified against other kit you could spend that sort of money on. The ADF is also very inaccurate, with 30 degree errors (within a specific close range to the beacon, say between 3nm and 5nm) being common.