I have no specific issues with BS7671 althought to be fair I don't agree with all of it.
I doubt whether any of us who are capable of some thought agree with all of it, so you're not alone.
You have described the situation I am talking about quite well, One of the areas I'm not fully in favour of is unprotected cables ...
You may not be "fully in favour of it", but you presumably accept that, in the situation I described (when it is considered "not likely" that the load would result in an overload current), omission of overload protection is compliant with BS7671? However,as you go on to ask ...
Who makes the decision that the overload is unlikely? The appliance manufacturer or the lad who has learnt to house bash on a 2 week special training course?
I would say 'neither of those'. As with many things, the decision needs to be made by someone who has adequate knowledge and thinking power
and who also does not have the sort of potential 'vested interests' that I mentioned in relation to the manufacturer.
My guess is the appliance manufacturer who have put many hours of research and developement including overload testing and risk assessment will come up with their percentages and include in their instructions the requirements for installation based on their rigorous testing procedures based on standards possibly set by outside bodies.
No matter what the manufacturer may know and/or believe (on the basis of testing or anything else),I think one nevertheless has to suspect that what iswritten in MI's may relate to(or be influenced by) those 'vested interests' rather than anything 'technical'. Don't forget what the Technical Dept guy at MK told me - that he could not answer for (and did not even fully understand) what was written in "Technical Data Sheets" or MIs, since they were written primarily by the marketing and legal departments, not the Technical Dept !
I'm not sure what you mean by "overload testing" but if a manufacturer really believed (on the basis of testing or anything else) that their product was "likely" to result in an overload current (the only situation in which BS7671 would require overload protection) I think it would be questionable as to whether they should be selling the product, wouldn't it?
Of course,if the flex supplying the oven were large enough to be adequately protected by a 32A OPD, then the situation would become even more straightforward, since BS7671 would then be satisfied by that protection
even if the oven were considered to be "likely to result in overload current", and regardless of anything the MIs might say!