... intrigued at the nature of your business which can brook no margin for public transport delays and manages to avoid delays on the road network.
Agreed, that wouldn't make sense. Which is why I was precise and specific in saying:
From my own records I know that driving most of the way down the M4 and then getting the tube is statistically both quicker and more reliably so door-to-door.
Risk of delays and disruption goes with either - the question is how
much risk. The records (which exist because it all has to be documented) tally with anecdotal perception - rail has proved significantly worse on this particular journey with a large enough sample to be confident in the conclusion. That's just a statement of fact.
Judicious use of Google maps with realtime crowdsourced data on this contra-tidal-flow journey along with intimate familiarity with all the one-junction diversion routes off the M4 and the likely consequences mean this is actually a really rather reliable journey at the +/- 10 minutes level.
Agreed. Crossing county lines? - not that. Delays by road - as unpredictable as rail delays. What is it then?
See above (and below). A county lines operation would be a good deal less legal but a good deal more lucrative!
Your post confused me: Bristol doesn't have a tube. Also, your complaint was about the reverse direction at the end of the afternoon. I assumed thus that your work was in Bristol and that you were on your way back. Any chance of elucidating what was so disastrous about being half an hour late home from work; annoying agreed.
To be clear, this is not a commute. This is a trip into London, often to the City or Southwark, timed to arrive soon after normal office hours end and returning on the penultimate or last trains of the day. I did this 83 times in 2019 - about 50% entirely by rail, the rest mainly by road, sometimes with tube or a bit of SWR. A pretty good sample then.
Obviously, if you have an important one-off meeting to attend somewhere, regardless of transport mode you allow lots of slack and absorb the waste of time as a necessary one-off cost. If you have to do something 80 times a year you don't - you regard a, say, 5% risk of failing to start on time as part of the deal. When that suddenly shoots up to 50%+, that's a problem.
I note that, since I stopped using it (!) the reliability of this service suddenly stabilised for a while. Before reverting to type...
All this waffle aside, I return to my basic question:
What is the actual point of a super-unreliable super-fast service? People who can afford to pad their schedule with a load of slack have no reason to care much about 10-15 minutes here and there. The people who do care (me - I'd love this service to run properly - I'd use it every single time if it did) paradoxically care precisely because 10-15 minutes actually matters. Blasting to London 15 minutes faster with a plan of... wasting an extra 15 minutes... makes no sense.