Latest Forecast: Labour Majority Reduced, Tories Static, Gains for Liberal Democrats

Note: The 2005 General Election has provided a great deal of data that will be used to improve future election forecasts - it was a much more complex election to predict than recent previous ones. This was primarily because vote swings differed from region to region, and from seat-type to seat-type, to a greater extent than previously seen. UK-Elect has a great deal of forecasting flexibility which can help cope with such factors, but still requires some judgement from the user to override default settings. In 2005 our judgement was a long way from perfect! We also significantly overestimated the late Liberal Democrat surge (and failed to account for the late success of the Labour campaign to frighten nervous anti-Blairites voters back to the fold with the threat of a Tory victory), leading to an over-estimation of Lib Dem seats and an underestimation of Tory ones. The analysis of what went right and wrong, and preparation for forecasting 2009/10 begins now...see our latest forecast for the first "for fun" effort.

Here's what we forecast back on May 5th..(Note that the "What if" links on this page still link to pre-2005 election versions which will be updated to their 2005+ equivalents over the coming weeks and months).

UK-Elect Forecast, May 3rd 2005. This is the latest and final  UK-Elect forecast for the next General Election, and represents what we expect to happen on May 5th.  Previous UK-Elect forecasts of this type have been quite accurate. Those before the 2001 election were within 2 seats for all major parties. (That said, the current election seems a difficult one to predict, as swings appear to differ between different seat types in different areas of the country.)

As with the April 17th Forecast, April 3rd/10th ForecastMarch 20th forecast, the March 5th forecast and the Feb 2005 forecast this forecast is again based on the new Scottish parliamentary boundaries.

This forecast assumes a lower Tory vote than most previous ones.  This is due to a feeling that the Conservative election campaign has faltered recently. This forecast  also assumes a slightly higher Liberal Democrat vote than the last few predictions. Our belief is that the Labour vote has reduced since 2001, however, it still looks more than sufficient to protect a large part of their overall majority. This is helped by a bias in the electoral system so strong that Tony Blair could win a majority even if the Labour Party came third in the polls. See "What if the Tories overtook Labour in votes?" What if the Liberal Democrat vote surged?, What if the Labour vote slumped?, What if the Tories reach 40%?, What if Tactical Voting explodes? and What if main Parties get equal votes? See also maps showing the 2001 General Election. and books such as From Votes to Seats: The Operation of the UK Electoral System since 1945

The percentages used for this forecast were Lab 37%, Con 32.5%, Lib Dem 24%. Other parties votes were not specifically set.  (If you prefer to forecast using your own percentages, the current opinion polls, or with a lower/higher level of tactical voting, download a trial version or browse the UK-Elect on-line shop. If you prefer to investigate what the betting companies believe the election results will be then click on the IG Index link to the left (hit your browser refresh button - usually F5 - if necessary to refresh it to show their current forecast majority).  If you believe in opinion polls too much, you probably need to see Opinion Polls: History, Theory and Practice or even Damned Lies and Statistics: Untangling Numbers from the Media, Politicians and Activists And if any of that is too heavy, you probably just need to have fun, in which case Prime Minister Forever is the obvious solution. Here's our review of this excellent General Election 2005 campaign simulation game)

Also, if you would like to forecast the Scottish Parliament, Welsh Assembly, London Assembly or your own local election, as well as the General Election, then click here to purchase

In this election simulation we have considered the current situation carefully, and have as a result slightly adjusted our forecasting parameters. This time we  assume that a maximum of 10% of people would be prepared to vote tactically against Labour in the right circumstances. (It should be emphasised that this figure is not the number that we think will vote tactically, simply the maximum number who would be prepared to do so if they lived in a constituency where the circumstances were right.) This number is a little lower than the last few forecasts, as we believe that the campaigns of the two major parties may have emphasised differences between the parties and dissuaded some left/centre potential tactical voters from opposing Labour as a protest vote. Because of this factor, the tactical voting calculation is also now adjusted to take more account of party second-preferences (e.g. both Labour and Conservative supporters are more likely to be prepared to vote tactically for the Liberal Democrats than for each other.)

Taking all these factors into account, the current official UK-Elect forecast for General Election 2005  is:



Change from current situation  Change from adjusted current situation  Change from 2001 election
Labour 374 -36 -25 -38
Conservative 172 +11 +8 +6
Liberal Democrat 69 +14 +15 +17
Scottish National Party 6 +1 +2 +1
Ulster Unionist 6 - - -
DUP 5 -1 -1 -
Plaid Cymru 5 +1 +1 +1
Sinn Fein 4 - - -
SDLP 3 - - -
Ind 1 - - -
Speaker 1 - - -
Labour Majority 103 -60 -55 -64
Forecast for South West England Forecast for Scotland  
Forecast for Wales Forecast for UK  
Forecast for London Forecast Gains  

Click on image to enlarge

Note: Labour majority shown simply excludes Speaker from calculation - if Speaker is treated as opposition MP then majority is 102.

With every forecast until the General Election we have also featured the detailed forecast for one constituency.

For this Featured Constituency Forecast we return to the Conservative / Liberal Democrat marginal of Dorset West:

Forecast for Dorset West, Dorset [South West England Euro/PR Region]
Electorate 74016

Justine McGuinness, LDm 23956 48.32%
Oliver Letwin, Con 21612 43.59%
Dave Roberts, Lab 3865 7.79%
Others 138 0.27%
Total 49571 66.97%
LDm Majority 2,344 4.73%

Liberal Democrat Gain From Conservative.  Shadow Chancellor Oliver Letwin loses his seat.
Swing from Conservative to Lib. Dem. of 3.67%

Note: This forecast is based on national trends only and does not take account of local factors, or specific other parties (UKIP and Green) standing in this seat that did not contest it in the last election..

Notes: A uniform percentage swing method was used, and the percentages were applied to GB constituencies only. The assumption was made that up to 10% of the electorate would be prepared to consider voting tactically against the Labour Party (Note to UK-Elect users: the simulation was set to take account of party vote transfer preferences, and TV percentage was therefore set at 20% as the maximum vote transfer for any party was set to 50%. Vote transfer preference for Lab and Con was set to 50%/10% to Lib Dem and the other main party respectively, whereas for Lib Dem it was set to 30% for both Lab and Con (a change from the default setting).  The number of iterations was set to 3.) The forecast base was the 2001 General Election constituency situation, adjusted to take account of the new Scottish parliamentary boundaries. No attempt was made to apply more accurate Scottish or Welsh poll percentages to the forecasting of constituencies in those areas (hence the forecasting of Nationalist gains or losses is likely to be less accurate). Results from Northern Ireland are based on those of the last election and included for completeness only. The "change from adjusted current situation" column represents the change from the  the current situation if the latest boundary changes are taken into consideration. The Labour majority figure here simply excludes the speaker from the calculation (Note: normally, in the event of a tie, the Speaker will vote with the Government.)

What if the Tories overtook Labour in votes?  What if the Liberal Democrat vote surged?  What if the Labour vote slumped?  What if the Tories reach 40%?  What if Tactical Voting explodes? What if main Parties get equal votes?




© Copyright 2006 UK Elect | About |