Strong Transition Model
Description of Electoral Calculus' advanced prediction model, which builds on the basic
transition model, but with extra features to incorporate incumbency and local party strength. The model
has been particularly designed to cope with weakened parties, such as the Lib Dems in autumn 2007.
Election History and Commentary
Overview of all election results from 1900, along with some commentary on elections
since 1997 and comments on how to use the site.
Liberal Democrat wipeout?
Predictions in October 2007 show the Lib Dems with no seats, which is obviously a bit low.
This analysis looks at why the model says this, the history of predicting Lib Dem results, and why the
outlook for them looks bad under any model
Gap analysis
Even when Labour and the Conservatives have equal support, Labour still wins about 90 more seats than
the Conservatives. Our new gap analysis explains the causes of this "gap", considers its fairness and
predicts whether it will last.
New boundaries for England and Wales 2006
The Boundary Commissions for England and Wales have redrawn many of the parliamentary seat boundaries.
This analysis gives an overview of the political impact of these changes, along with lists of the seats which
are disappearing and the new seats which are created.
Track Record - 2005 Election
Comparison of the actual 2005 election result against our prediction. What did we get wrong, and what
were the cause of the errors.
Scenarios for 2005 Election
In the run-up to the 2005 election, we analyse likely scenarios by examining the final opinion polls,
betting markets, Electoral Calculus users' trends, and tactical voting.
Track Record - 2001 Election
Comparison of the actual 2001 election result against our prediction. What did we get wrong, and what
were the cause of the errors.
Tactical Voting Model
A description of our tactical voting model, along with numerical evidence from the last four election results.
We focus on the differences between the actual result and our basic prediction, and see how much of the
difference can be explained by tactical factors.
Science of Opinion Polls
A brief note on the science of opinion polls. How much error can be expected due just to random chance?
Transition Model for Predictions
A full description of Electoral Calculus' basic transition model for predicting elections.
The new model avoids some problems of the additive swing model, whilst retaining its good properties,
and making very similar predictions for which party will actually win each seat. Backtesting each model
against the 1987-2001 elections showed quite good performance from both, with the Transition Model being
slightly, but not significantly, more accurate.
Track Record - Overview
Summary track record of Electoral Calculus predictions against actual election results for
1992 to date.