While I generally agree, there should be some gaps. As this is the only realistic way that maintenance staff can get in at busy junctions if there is a failure. Unless you want every single train going over the junction to be delayed until the middle of the night…
Engineering of junctions and other infrastructure should avoid single points of failure and generally be engineered to reduce the number of expensive shutdowns.
However, if there is a failure bad enough to stop the job, surely the staff could access it immediatley because there is no traffic preventing them doing so?
Otherwise, a delayed train is infinitely preferable to no train, which is what you would have if you start carving chunks out of the timetable.
However, a metro-ised timetable would tend to have less need for very complex junction infrastructure, since the timetable would necessarily require a comparatively simple set of movements through the junction
EDIT:
And if things are that bad, you can just kill the service through the junction temporarily and advise on alternate routings. Which is far more possible in an all clockface system.
Isn’t that your metro-isation?
Have you not noticed that with less staff, when it does go wrong, it goes far more pear shaped than if you have some actual experienced staff around to help sort things out? Plus, with the general population getting older, and the requirements to help the disabled, it’s preferable to not cut staff.
Staff are expensive, incredibly expensive.
Tolerating occasional failures is preferable to spending vast sums maintaining staff to possibly mitigate some of those problems.
This will become especially true as, as you say, the general population becomes older and less fit. Labour costs are already high and are going to go through the roof.
And ofcourse employing large numbers of staff comes with its own set of problems.