Is no one going to argue with my assertion that the 09:37 Oxford could do the Oxford to Padd journey in 30 minutes less than currently scheduled?
I'll throw in my expert opinion
30 minutes would be pushing it a little, but I reckon 20 minutes quicker might be possible. The layovers on most of the stoppers (in the up direction) are around 7-12 minutes at Didcot (depending if it's made any stops at Radley, Culham or Appleford), and 5-6 minutes at Reading. The timings are a little slack from Reading to Ealing Broadway, maybe a couple of minutes could be trimmed, and then there's another 3 or so minutes of slack between Ealing and Paddington - usually whilst awaiting a path into platform 11 whilst the xx:27 leaves Paddington. So, that make a maximum of 23 minutes by my experience and calculations.
Trouble is these freight trains get in the way all the while, and when the stoppers were timed to do the trip much quicker back in 2006/7 - the down ones were timed for 1h 25m with stops at Ealing, Hayes, Slough, Maidenhead, Twyford, Reading, Tilehurst, Pangbourne, Goring, Cholsey, and Didcot - there were quite often delays of 5 or so minutes, which would have had a marked difference on the PPM figure. The usual reason for delay was either catching up with the previous stopper (which in those days made more stops) between Slough and Reading, or waiting a liner to come off of the Reading avoider and clear the long signal sections as far as Pangbourne. With no recovery time at all, it was then game over for making the PPM.
That being said, there really should be a more attractive timetable devised with most of the slack at Didcot and Reading removed, and a journey time of around 15 minutes quicker than present achievable with a certain amount of slack still allowed for. Trouble is, that's easier said than done with XC and freight trains in the mix on the relief lines.
The 09:37 was on time today though!