OK, but it's not the terminology which is the issue but, rather, the concept - there is a big difference between a moderate degree of overload (everyday sense) which is technically called 'overload current' and a very large degree of overload (everyday sense), which is technically called 'fault current'. As you have described, in the former case it's not impossible that the current will be greater than the CCC of the cable but less than the rating of the OPD, but in the latter case it's inevitable that the current will be greater than both the CCC of the cable and the rating of the OPD.
We're down to probabilities (likelihood) again. I would say not unreasonably, the regs love the word "likely". The situation you describe would only be "within the regs" if, in the judgement of the designer, the load was not likely to (under fault conditions) draw a current in excess of the CCC of the cable. The existence of anecdotes (such as you are able to provide) does not necessarily mean that such events are "likely".
Kind Regards, John