Fair enough on a dedicated basis. But when I briefly served with BREL, 1974-5, we were told, by a BR Board member for engineering no less, that the Deltics' annual maintenance costs were approximately double those of the next expensive class (which were the 'Western' hydraulics) and around four times the costs for the like of Brush 47's. Part of that no doubt is due to lack of economies of scale - 22 locomotives (nearly said 22 engines but of course in the Delts that would be 44) of unique design, as opposed to spares provision for around 500 locomotives (and those with some commonalities with other classes).
So as long as the Delts ruled on the ECML this was, presumably, factored into the ticket price - but never any economic case for cascading onto less profitable secondary services, except, as in fact happened, just to use up the mileage before the next general overhaul.
I imagine the equation was somewhat similar for Princess/Coronations, Kings (with the other problem of limited route availability) etc. Curiously, some of the LNER pacifics do seem to have been slightly (but only slightly) more versatile in achieving a 'second career'.
This is a contrast (and possibly to the railway accountants a shock) if we compare it with the previous generations of 'express passenger' engines - NER Atlantics, GWR Stars, MR Compound or CR Pickersgill 4-4-0s, and doubtless many others, which did have a properly economic utility on secondary services after their glory days were over. I don't think this is just because of diseaselisation - the big 1930s-1950s express engines, or the Deltics, were simply never going to be viable if downgraded to, say, Hull-Leeds semi-fasts, or stoppers on the ECML.
Which makes me think of Thompson. How far, in the depths of WW11, could he even dream of restoring or improving the elite services of pre-war (which as a previous poster has noted, may have been as much about PR as real revenue)? It seems to me that a lot of what he was trying to do was around reliability and standardisation - OK some 30 years after Churchward. Cylinders, boilers (also of course incorporating the potential for newer manufacturing methods: you can now roll a barrel in two rings rather than three, so why wouldn't you.?) Arguably he went too far, too soon, at a time when innovation, even if it might ultimately prove to be the right path, was in the short term more costly/disruptive than 'business as usual', there being a war on and all that.
He had a strong North Eastern legacy, and that railway arguably had, even more than the GWR, a long tradition of standardisation - the Worsdell brothers and then Raven really tried not to introduce a new boiler diagram or a new wheel diameter or a new cylinder block design if they didn't have to, and also were famous for 'rebuilding' (one uses the term loosely and sometimes in accountancy terms) earlier locomotives to be for practical purposes identical to their new build successors. On the other hand, the NER tradition may have blinded him sometimes: Raven was keen on monobloc three-cylinder castings, which doesn't seem to have mattered. On Gresley designs, especially the V2, the monobloc was a weakness - one cylinder gone, you have to replace all three. (Off topic but you can argue the same with articulated train sets from the Silver Jubilee to the APT - an elegant solution, until something goes wrong).
Plus side, the monobloc cylinder casting is, as far as I know, the 'scientific' justification for the preservation of Green Arrow as part of the national collection - and also, consequentially, why she may never be returned to steam.
I trust readers will forgive these musings: like many other posters, if I assert something, it's really a request for more knowledgeable folks to show me I'm wrong!