You might be wondering why the Sunday papers were filled with conflicting version of the results from the municipal election.
The answer is contained in a decent story on Times Live written by Brendan Boyle:
The DA took 23.80% of the vote for ward candidates, 24.08% of the proportional representation vote for parties, 15.3% of the vote for ward councillors and 21.97% of all votes cast.
The ANC took 60.98% of the ward vote, 62.93% of the proportional representation vote, 69.43% of the district council vote and 63.65% of all votes cast.
So you can spin your version in a number of ways – and everyone has been furiously doing just that. See here for the DA using the proportional vote comparison which gives it 24.08 percent versus the ANC 62.93 percent and therefore casts its performance in the best possible light (something it is perfectly in its rights to do.)
(This added half an hour after publishing – it has been pointed out to me by some of the people I follow on Twitter – @Bruceps and @RyanCoetzee – that the only choice that all voters were offered that indicated their party preference is the proportional representation vote which gives the ANC 62.93 percent and the DA 24.08 – strongly persuasive that this is the number that most accurately indicates party support.)
The Sunday papers seems to have arbitrarily shifted from one usage to another.
I have used the figures from the IEC website for the overall total of votes cast for parties. Here they are (if you click on any one of the nine it will enlarge and become readable):
Nic Borain 
Politics.
Poly = many, tics = blood suckers. Politics = many blood suckers.
http://jahtruth.net/britmon.htm#crimes
Very informative and accessible, thanks Nic.
Don’t be polite to me Andre, I rely on you to kak me out when I soft-soap the global corporates or generally engage in liberal waffling – and any other temptations of the flesh
At last something we can refer to. Thanks Nic
So, as I understand it, urban voters had 2 ballots, and rural voters had 3. Simply totalling all ANC votes across all ballots and dividing by total number of ballots gives ~64% ANC support across all ballots. However, this simple arithmetic is slightly misleading, since ANC support is higher in rural areas. A balanced calculation (/3 for rural votes and /2 for urban votes) gives ~62% support for ANC nationally.
But, for the DA this election was all about the big metropoles: Cape Town, Nelson Mandela, Tshwane, Joburg. In all four significant gains were made by the DA, but largely due to consolidation of minority party votes. Nevertheless, ANC support in Tshwane and Joburg dropped by a few % points against both the last municipal and general elections. Interesting, cos it’s already close to a tipping point (ANC support in the 50%’s).
So, unfortunately, next election might be the more interesting statistically 😉 Or not.
I’m with you on this … you wont believe how long it took me to work out the fact that your first para records