If Central London had a Mayor, Ken Livingstone would win by a landslide. But in the 1970s the Tories expanded the political boundaries into leafy outer London.
Since then (like Scotland & Northern England) Inner London has had to endure Tory rule they never voted for. Would Inner London have been willing to tolerate the transport chaos and pollution on its doorstep (which only benefits commuters from Outer London) without this change?
Back in the 1950s the Tories could still win the majority of seats in Scotland. Which today seems incredible as the Tories struggle to hold even one seat there. And right into the 1980s, Tories could win in the urban North. But Thatcher put paid to that.
And it is quite conceivable that Northern England could become like Scotland with hardly any Tory MPs if present trends continue over the next few decades. Just 13% of Tory MPs represent constituencies in the North of England despite it being 24% of the UK electorate. And none of those represent the urban areas of the big cities.
Saying that, the Tory decline in the North is not as bad as Labour's decline in the South outside London. Only 3% of Labour MPs currently represent constituencies there despite this area also being 24% of the UK electorate. Something strange is going on that goes beyond sheer economics or ideology.
This also explains why the Lib Dems strength is crucial to Labour beating the Tories. The Lib Dems win 34% of their seats in the South. 19 seats to Labour's 8. These 19 seats would all go to the Tories if the Lib Dems do badly - which currently looks likely with low polling and radically changed boundaries.
All over the world, the Right has always done better in rural and small town areas and the Left in urban centres. But our voting system seems to exagerate and accelerate this process. Why?
I think part of the answer lies in the fact that people like winners and our winner takes all system makes change very difficult. Constituencies and councils can remain one party hegemonies for decades or longer.
In fact big change only really comes with significant boundary shifts like happened for the 1983 Tory landslide or 1997 Labour landslide that gave both parties long periods in power on ever diminishing minority support.
It took nearly 70% of opposition votes to finally remove the Tories and over 70% for New Labour to lose, and only then to a Tory/Lib Dem coalition.
In short, our system is hopeless at allowing voters to 'throw the buggers out'. The bigger power lies with the boundary commissions.
Once a party gets significantly ahead in a seat and national swings have no effect on the result there, it becomes a self fulfilling prophecy as opposition voters give up and can't see the point in turning out to vote for someone who can't win.
And the urban poor reach a point of despair quicker than the rich do in rural areas. This bigger drop in turnout in safe urban areas over the smaller drop in safe seats in rural areas is something the Tories are seeking to exploit by the new proposed enlarged boundaries and changes to discourage electoral registration.
The new boundaries are the Tories way of getting round the inconvenient problem of not getting enough votes to govern. The new boundaries will open the door to the big two parties gaining power on even fewer votes.
Sadly it is inevitable the gerrymandering will reach US proportions in the not too distant future. There, the governing party use computer programmes to draw boundaries to maximise the number of seats they get. There is little pretence about this and districts just have a number as there is no correlation with actual communities.
The frequent changes in boundaries before every election will hand more importance to the media as hard working MPs find their electoral base continually disappearing into neighbouring constituencies thereby making local campaigning less significant. As the media generally backs the Tories it is easy to see how fluid boundary changes helps them.
But even continual churn in boundaries might not be enough for the Tories.
By 2020 the Tories will also gain from the new registration rules disenfranchising more Labour voters and making urban Labour seats bigger, but it still leaves the regional problem of lack of support in the North.
The next few elections are crucial - the Tories need more than 36% of the vote nationally and it seems unlikely to get it, especially in the North as public sector redundancies devastate communities like mining redundancies did in the 80s.
There is another problem for the Tories that the US doesn't have. Plurality of parties. In the US a duopoly restricts competition through tough rules on who can stand. It is also much more prohibitively expensive to put up candidates than in the UK. Only the winner takes all system holds back new parties here but that is still one hell of an obstacle.
It is hard to overstate just how important boundary changes are to results. The common misperception with bias in our system is that constituencies need to have different size electorates as there were in the 'rotten boroughs' era but this is simply not true.
Also, most assume bias is deliberate, but even with identical size constituences and completely impartial drawing of boundaries the system can still throw up some quite shocking bias, as demonstrated by the 'gerrymander wheel' - http://www.stvaction.org.uk/gerrymander_wheel
You have identical voting patterns, but by just changing the boundaries you can vary results from a landslide for one party to a win for an opposition with far fewer votes.
We can see how this happens in practice by looking at the new boundary proposals for Brighton Pavilion, Brighton Kemptown, Hove, and Lewes. All currently marginal seats.
Currently, the 4 constituencies (all fairly equal in size) on current polling would almost certainly deliver MPs of 1 Green, 2 Labour and 1 Lib Dem as Labour support is up 11% nationally since 2010 and both Green Caroline Lucas and Lib Norman Baker have strong backing in their respective constituencies to hold on.
But the 3 new constituencies to replace them - Brighton Pavilion & Hove South, Brighton & Hove North, and Lewes & East Brighton will most likely deliver 1 Green and 2 Tory MPs.
The voters vote the same in both situations but the new Pavilion seat has been made a very safe Green seat and the combining of Lewes and Brighton has split Lib and Lab support. Ward results in local elections show that by doing this, 2 safe-ish Tory seats are created.
It would have been possible to have kept 2 of the 3 seats as marginals by drawing boundaries that actually make more sense than has been proposed.
Basically the new proposals slice central and west Brighton and all of Hove horizontally except mysteriously for a constituency called Brighton Pavilion and Hove South they put 2 coastal wards from Hove south in the more northern constituency and include 2 northerly wards instead.
If this was reversed it would still keep the constituencies the same size and would create 2 Tory/Green marginals instead of 1 safe Tory and 1 safe Green seat.
Why the boundary commission of England have favoured the Tories in this way I leave you to consider (since I believe voters given a choice would choose the Greens over the Tories in both seats).
And this seems to be the pattern the commission have followed right across urban England - creating more Tory safeish seats at the expense of Tory/Labour marginals that Labour would have won.
I think the pay off for the Tories will be bigger than the 20 seats being predicted. We could get a Tory majority government even though their vote falls. The wonder of boundary changes would be demonstrated. Sadly most voters know little of how they are being conned.