The NEP analysis of the 2004 election exit polls states (on page 5):
“It is important to note that the exit poll estimates did not lead to a single incorrect NEP winner projection on election night.” .....hang on a minute. Can that be right?
First take a look at their figures on election night, all with links to the supporting evidence.
- National exit poll (updated at 7.38 p.m) Kerry 51.5% Bush 48.5%
- National exit poll (updated at 12.23 a.m) Kerry 51.3% Bush 48.7%
(for clarity the figures are the percentages of the total vote excluding other candidates)
Then there are the 4 states where NEP put Kerry in the lead - but which all went to Bush!
- Iowa exit poll (updated at 12.23 a.m.) Kerry 50.7% Bush 49.3%
- New Mexico exit poll (updated at 12.21 a.m.) Kerry 51.3% Bush 48.7%
- Nevada exit poll (updated at 12.23 a.m.) Kerry 50.7% Bush 49.3%
Ohio exit poll (updated at 12.23 a.m.) Kerry 52.1% Bush 47.9%
So how can they say:
“ the exit poll estimates did not lead to a single incorrect NEP winner projection on election night.” ?
Simple - it’s called newspeak. NEP states that they don’t predict a winner until they have adjusted the exit polls to the actual results. In other words all the above predictions were changed when they saw they were wrong, so therefore all their predictions were right! In fact why not quote a margin of error of 0% –which, using that method, could be achieved by any fool.
We nearly forgot to mention Florida, where NEP predicted a dead heat (www.exitpollz.org/pages/florida.html)
And while you’re there check out the prediction for Miami taken from the full Florida exit poll. It shows a 9% discrepancy compared to the declared result for Miami-Dade county! We are still waiting for the explanation.
You could try looking in The NEP report (pdf 77 pages) -or you might, judging by the above statement, decide to try looking somewhere else!