In reply to Bellie:
It's a wasted vote for those in a safe seat whose vote will never contribute to their party's representation.
It's lunchtime so had a play with Excel...
In the 2015 election there were 360 constituencies with majorities of 10,000 or more. In those constituencies 7.9 million people voted for candidates who lost and 9.5 million for the candidates who won.
30.7 million total in all 650 constitutes voted but 7.9 million of them (in 360 safe seats) had no chance of actually electing anybody of the party of their choice.
If you extend that to include all the 505 constituencies with majorities of 5000 or more then 11.5 million people voted with no realistic chance of contributing to their party parliament.
That's not much less than the 15.3 million people who voted for the 650 candidates who won.
I don't think safe seats are a good thing as they breed political complacency or a quite logical apathy among voters.
I guess 80% of us live in the ~500 constituencies where you pretty much know which party will win.
FPTP does work* but it seems to be one of the most imperfect versions of the many imperfect versions of democracy.
There's too many people who can vote with no prospect of electing anybody or contributing to additional representation when they have choice between no-hopers or boosting a near guaranteed majority in a safe seat.
*although a result of SNP 56 out of 59 Scottish constituencies can't really be described as "working".
You can get the numbers to play with at
http://www.electoralcommission.org.uk/our-work/our-research/electoral-data
Post edited at 14:28