Indeed, there are 2 issues:
(1) can the IEPs match HST performance on diesel, on routes planned to be wired
and (2) can DfT/GWR afford the increased leasing costs for doing so, because of the higher maintenance/lifetime costs
A Yes answer to (1) doesn't necessarily also mean Yes to (2).
They are already due to pay higher leasing charges for the electric IEPs converted to bi-mode.
A few thoughts:
a) The IEP Dwell time should be lower due to non manual doors and the timetabling risk that door(s) isn't/aren't closed.
b) IEP acceleration and braking even at original spec on diesel should be far better than HST.
So that might produce a higher performance without needing too much time at higher rated engine power output (and hence fuel consumption) to compensate for the less than 125mph max speed. Especially if lots of running at less than 125mph and lots of stop the top speed matters less. The via Oxford services would have virtually all the high speed running under electric so might make the most sense in the circumstances if worrying about fuel levels as there would be the least difference overall in terms of extra Diesel running.
c) I haven't been following the GW electrification ultra closely but I assume that the 2 main issues that will caused delays on the Didcot - Bristol Parkway route are at Steventon then the Cotswold AONB section resulting in the December 2018 go live date with the other parts able to be completed far sooner, the question is could they be energised sooner as well? E.g. current piling in Wooton Basset Area but theoretical go live in 22 months...
Similar situation for Bristol Parkway - Cardiff presumably a useful section could be energised earlier than Decmeber 2018 given the amount of work completed to date (piling, bridge rebuilding, resignalling)
So would another section or 2 available earlier reduce a sensible amount of diesel running in terms of refuelling and diagramming?