The problem with marginal schemes is that they take personnel and plant away from other projects. That admittedly less of an issue with 750V DC schemes but they do still require specialist staff, and there's usually some sort of third-rail power supply upgrade or replacement at some stage of development or implementation (and not just around London).
The point I was making wasn't that we should be prioritising marginal schemes over more worthwhile schemes.
Rather that some of the schemes which would otherwise have been discounted will, in time, probably actually get built.
We're already at the point where rail as a whole (i.e. including diesels) is already marginally better than electric cars in terms of emissions on a traction only basis (life cycle is been ahead for some time) and there's a lot more which can be done a lot more easily to get the revisions down further and faster than electric cars can.
One of the big things is more electrification (of any type) and probably part of the shift towards considering more third rail is perhaps a bigger picture view. In that maybe the question is being asked more along the lines of "whilst there is an increased risk of harm from third rail, how does that compare to the level of harm from not getting our emissions down?"
Perhaps with the follow up question "if we allow some infill schemes which are already in third rail areas, where there's limited (ideally no) extra powered depots required and ideally in areas where there's generally less risk of people ending up on the lines, but with a plan for reducing third rail infrastructure slightly longer term, then could the risks be acceptable?"
In that (and this is mostly guesswork on my part, so feel free to correct me if I'm wrong) in my head the main risk to staff day to day is in the depots where drivers are walking about to get to/from trains
Now whilst there's a risk to maintenance staff, with a lot of work done overnight and within blockades that is reduced. If this was also paired with more restrictive working patterns and extra protection built in the design stage (such as being able to isolate fairly short sections track, having frequent access points to the track, extending working hours so that works can be done more easily overnight with the juice turned off). Especially given that many infill schemes are likely to be on lesser used routes and so wouldn't be subject to the level of maintenance of the lines within the M25.
Add to that the potential for not being in urban areas with some being through fairly rural sections, and therefore the risk of trespass is reduced.
As such although this rail may well be (say) 120 vs 100 for OHLE over a given length with the above lower risks it may fall to an acceptable level. Now whilst OHLE may then also fall on the same measure I suspect that the gap may well have closed a bit.
As such with maybe a decade before any extra third rail installs and then maybe another decade before third rail starts to be replaced by OHLE it could well be that there's only a short period where the overall milage of third rail is (say) 3% more than the current figure before falling to 2% less over the following decade.
To put some numbers to that, there's currently 4,285km of third rail. An extra 1% would be an extra 43 miles, so 3% would be about 130 miles. Without looking at all the schemes which would make sense, that's probably about the right sort of ballpark to do the key infill schemes. (2030 to 2040).
That then would require about 90 miles more to be converted (220 miles) over the following two decades to then got back to the 2% below (2050 onwards). It may well be that it extends beyond that by a decade, but that kind of timeframe.
As such much of the urgent (i.e. to get it done to be net carbon neutral by 2050) should have happened and the conversion to OHLE could then run in parallel with the revival of battery trains.
In doing so there shouldn't be as many "why is electric rail getting replaced when so many other routes need to be electrified" as many of the key routes should be electric anyway, or at the very least battery trains.
Having such a milage limit (even if it's just a goal to keep it to that miles rather than a actual limit) would however mean that it would only be infill schemes. As such projects like Basingstoke to Southampton via Salisbury, and probably to Yeovil as well, would then be OHLE. Even Reading to Gatwick would likely be OHLE from near Guildford to Dorking.
Then the bits which would then likely be converted first would be:
Basingstoke - Southampton
Southampton - Weymouth
Basingstoke - Farnborough
Wokingham - Aldershot
Alton - Aldershot
Ascot - Aldershot
Farnborough - Woking - Guildford
Guildford - Aldershot
Guildford - near Havant
Havant - Southampton
Havant - Portsmouth
Havant - Chichester
Reading - Bracknell
Probably several of them would happen in parallel and there's likely to be other lines in other areas which would also happen, such as from Brighton inwards and each way along the coast and probably some of the lines in Kent as they connect to or are near to HS1.
Brighton could be a good one, as a lot of the trains are already dual voltage due to running through the Thameslink tunnels.