A brief aside: UK by-elections

I'm preparing a briefing on the battleground states for the US presidential elections for Monday.
Meanwhile in the UK, two by-elections to replace Labour Party MPs have resulted in a win for the third-party Liberal Democrats in Leicester South and a narrow win for Labour in Birmingham Hodge Hill.
A lot of nonsense and spin will be written about these results over the next few days but the picture is very clear.
The Conservative Party has not gained a House of Commons seat from a parliamentary by-election since June 3 1982 (Mitcham and Morden in SW London). From November 1990 until May 1997, the Tories were unable to even hold any of their own seats in by-elections, no matter what the majority.
Quite simply, a party that cannot make even a single by-election gain when in opposition, is not going to win a general election, especially as the timing is decided by the Prime Minister Tony Blair. The power to choose the timing of a general election is arguably the most powerful political weapon in British politics: it can even help to keep rebellious members of the Prime Minister's party in line.
Far from challenging Labour as the party of government, the Conservatives are struggling to establish themselves as even a vehicle for protest votes.

Submerged with data

Thursday's American Enterprise Institute Election Watch briefing was especially informative, dealing with the "bounce" in the polls that coincides with vice-presidential nominations and the party conventions. Check out the handouts for reams of figures as well as the video.
The polling data all points to a close contest to be won by Kerry. What I am not seeing is how this compares with 2000. We should remember that the opinion polls at the equivalent stage predicted a Democrat victory for Al Gore. Intriguing fact of the day, the Gallup polling is in 16 states, nine of which were held by Gore in 2000. Is Bush really threatening to win Michigan?
I will report on this further, but the thought occurs that we could be looking at a 1987 British election where the BBC announced a hung-Parliament at the close of polls only for a 100 seat Conservative majority to be declared when the actual votes were counted.
Ignore the swings in the polls before the end of the Republican Convention in late August: Kerry isn't likely to show better poll ratings before then than Michael Dukakis did in 1988. By November, Dukakis held 11 states to George Bush MkI's 40.


The weight of history or chance?

Because the boundaries of the U.S.A.'s individual states don't move as much as English local constituencies do, it is possible to measure election results over quite long periods of time and pick out certain themes.

For instance, there are six states that every winning candidate since 1972 has won, out of a possible 50 plus Washington D.C.. These are Arkansas, Kentucky, Louisiana, Missouri, Ohio and Tennessee. Not surprisingly four of these states are among the biggest Democrat targets for this year's presidential election.

The state which is most remarkable for swaying with the prevailing political direction is Missouri. This Mid-Western state was won by George W. Bush in 2000 with a majority of 78,786 or 3.24%. With 11 electoral college votes, Bush could not afford to lose Missouri without making gains elsewhere.

However, if we suppose that President Bush could be elected without Missouri in 2004, he would be only the second Republican since 1900 to do so. The other overall winner without Missouri was General Eisenhower in 1956 (the year he was re-elected).

For the Democrats to win without Missouri would be even more historic. The last time they did so was in 1916 with Woodrow Wilson. That was also a re-election. There have been Republicans elected to the presidency in the XIXth century. But since 1856, (the first Democrat versus Republican presidential contest) no Democrat apart from Woodrow Wilson has won without Missouri.
[Updated November 22 2004]

Because Missouri is also hosting an election for (currently Democrat) Governor, one of its (currently Republican) Senators, and the Members of the House of Representatives (currently 4 Democrats and 5 Republicans), this is my pick for the single most interesting contest of the campaign.


Bookies and in-depth poll suggest that Bush is underestimated... again

Apart from the remark that an opinion poll with a three and a half per cent margin of error is pathetic this poll reveals opinions that seem likely to favour the incumbent at the polling station.

Let's be clear just how poor this polling is: you could predict that Ralph Nader would have got zero votes in 2000, or that Bush would have won five extra states.

I've said on Samizdata that when it comes to forecasting, I trust bookies better than pollsters: we're talking people putting their money where their mouth is, and gamblers are not spin doctors. Wired shows the gap between people voting with money rather than with gratuitous chat.

There have been rumours that some unscrupulous individuals weight opinion poll samples to put more registered voters from a certain political party. As if by magic, this produces the result the media outlet wishes to peddle.

With a U.S. election I'm afraid polling is bound to be tricky. One million extra Democrat voters in California compared with last time don't mean a thing. Six thousand more Republicans in Wisconsin mean a lot more. Now design me an opinion poll which factors these issues in a country with relatively high voter mobility.

My preferred approach is to pick out the swing states and check them out with a mixture of polls, random interviews, even check out the TV advertising schedules.


Welcome to Antoine Clarke's Election Watch

I've been gathering information and analysis for the November 2004 elections in the U.S. over the past three months. My track record in elections is generally good, especially long-range forecasting. I have "called" elections (mostly correctly) in the U.K., France, and the U.S. for the past 17 years.

I shall be uploading some of my material over the coming week and comment on new developments as I go along.

Part of my service will be to publish a campaign status: what the current outcome would be, given what predictable events are between now and election day. At this tentative stage, I put the presidency as marginally Bush, the Senate a Republican hold with an increased majority, the House of Representatives open but marginally Republican, and the Democrats to make no net gains in the governor stakes.

On Monday, I shall report on the importance to the various parties of the different campaigns.

I am also trying to get around the problem of spamming so for the time being please email me at antoine.clarke[at]gmail.com.