Note that during the first half of the eighties you could travel at up to 125mph in the leading cars of the APT-P trainsets in passenger service (And the trains themselves were designed for 155mph operation - I'm not sure whether they ever operated faster than 125 with anything other than test staff and crew on board).
The HSTs required two locomotives to achieve the required levels of performance, so it made sense to have one at each end of the train to create an efficient push-pull trainset. It wasn't possible to do this with the electric APT trainsets, due to concerns both over waves in the contact wire reducing the rear pantographs' contact with the wire, as well as with running a 25kV bus down the length of the train, so the two power cars required were located in the centre of the formation, necessarily splitting the passenger accommodation in two and duplicating staffing and facilities.
It's only speculation, but I suspect that transformer and traction motor technology wasn't sufficiently advanced in the seventies to allow for an AC electric multiple unit powerful enough to achieve more than 100mph within the limitations of the UK loading gauge: I note that the 0 series Shinkansens had a pantograph on every second vehicle - was this due to the maximum size of transformer achievable with 1960s technology even within the far more accomodating Japanese Shinkansen loading gauge?
In my view BRs' approach to train operation remained very steam age through the seventies, and attitudes only began to change with the introduction of the HSTs, but at the time putting a thwacking great lump of locomotive, with lots of space above the solebar for all the internal 'gubbins' and plenty of weight for adhesion, on the front of the train remained the most effective means of accelerating a long string of carriages up to a decent speed in good time and maintaining that over long distances, based on technological and space limitations. Plus it gave the flexibility to vary train lengths to suit seasonal demand.