The ECS runs I've had in mind are the ones in the LTV area. For instance:
1N40 16:09 PAD - MAI, arrives at MAI at 16:42, then departs as
3N40 MAI - PAD at 16:46.
It isn't going through non-passenger lines, and just chugs along the reliefs all the way to Paddington. It has a 4 min turnaround time at Maidenhead, so would be enough to pick up some passengers. Right behind it is a stopping service towards Paddington.
Also since the new timetable, due to the Marlow branch junction redesign, most through trains to Bourne End are terminating at Maidenhead, then running as ECS to either Paddington, or Reading. Another example:
1N44 17:42 PAD - MAI, with a 5 min turnaround time, then going back to Paddington as
3H62 18:19 MAI - PAD. Again, no non-passenger lines.
At Twyford, we have the same happening.
2H49 18:17 HOT - TWY mysteriously terminates at Twyford, then continuing as
3R77 18:33 TWY - PAD.
Later, we have this:
1N52 18:42 PAD - MAI, which normally terminated on Pl.3 at Maidenhead (unsure why it was routed onto P4 that day, maybe due to the signalling problem at Paddington causing delays to trains coming out of London), which is the platform that goes towards Reading anyways. It then turns into
5N52 19:18 MAI - RDG. This example I don't understand the most, since it doesn't need to be turned around, and continues to Reading anyways, but not in passenger service past Maidenhead.
Another example of a Henley-on-Thames - Twyford, then to Paddington as ECS:
2H53 19:18 HOT - TWY &
3P10 19:32 TWY - PAD
That concludes the PM Peak examples I have that caused me to start this thread.
From the view of a passenger, it's just strange that if I wanted to travel to London from Henley, I need to change at Twyford, even though the train continues onwards to Paddington anyway.