The lizards were right (up to a point)
You may or may not have seen http://www.plosone.org/article/info:doi/10.1371/journal.pone.0014248
research when it came out. Essentially they have taken a month's worth of anonymised BT phone records and mapped the call patterns in an attempt to see what regions emerge from the data. Obviously this is not going to be bad as a proxy for regional connections.
Now, amusing though the graphs are that partition the data into firm boundaries, I think the fact that the middle and right versions of figure 2 differ so much - essentially tweaking of the parameters causes the region based on the Bournemouth-Poole-Southampton-Portsmouth axis to disappear and be replaced by the one to the West of London - means that this should be taken with a pinch of salt, and that figure 1 which distinguishes between the strong connections and the weaker periphery is more interesting (*) - it also shows the current RDA / regional assembly areas. These have of course gone as a key belief of the ZaNuLieBoreGordonBrownGordonClownMoreLik
eists is that the regional assemblies were a front for the mind control lizards of Brussels.
So... surprising things
1. I wouldn't have guessed that the green zone South of London is a distinct region and that it goes so far North i.e. that the London zone goes further North than it does South. Otherwise I was actually surprised that the London zone is so small - there is essentially the Greater London area going North as far as Luton, and a hinterland with a weaker association
2. I was surprised, but maybe shouldn't have been that the Birmingham region goes as far West as Aberystwyth.
3. The Bristol / South West region goes further East than I would have expected (perhaps this shouldn't have surprised me either)
Things I would have guessed, I think
1. The East Midlands and Yorkshire and Humberside areas have the least cohesion (apart from a core West Yorkshire area), with the North-East and North-West / North Wales areas being more strongly delineated.
2. Oxfordshire is very much in an ill-defined area between the South West, London, Greater Birmingham and Greater Southampton (hence the random local news we end up with)
So, what are the deep and profound conclusions? Well, the main one that strikes me is that despite all the relentless bigging up, London's area of influence is surprisingly small. And, as I said before, maybe the money for HS2 should be spent on lines not going to/from London in the first instance.
(*) although it is amusing that the first partitioning which emerges is the split into Greater London, Scotland and the rest of England and Wales.
EDITED TO ADD: Also amusing, if you look at the left and middle maps in figure 2 you will see that Oxford appears as an outlier of Greater London (along with Basingstoke and somewhere weirdly between Bath and Chippenham)