The headline prediction for the May 2015 election was not accurate. The final prediction was for a hung parliament with Labour/SNP as the largest bloc. The actual result was a small Conservative majority.
In numerical terms, the prediction and the outcome were:
|Party||2010 Votes||2010 Seats||Pred Votes||Pred Seats||Actual Votes||Actual Seats||Vote Error||Seat Error|
The Conservative support was significantly underestimated, which caused the number of Conservative seats also to be underestimated. Although the Labour support figure was quite accurate, the error in the Conservatives caused the predicted number of Labour seats to be too high. The Liberal Democrats were also partly overestimated, but the prediction was accurate in saying that they would lose the vast majority of their seats.
The predictions for the other parties were relatively good, getting Plaid Cymru, UKIP and the Greens exactly right (including the exact seats held and gained), and being fairly accurate in predicting the landslide win for the SNP in Scotland.
In total sixty-four seats were mis-predicted. This would be a moderately good result if the general trend had been right, but the overall prediction quality was poor at this election. This was mostly due to polling error in the pre-election opinion polls.
We will now look at these and other issues in more detail. The particular topics studied are:
|Pollster||Sample dates||Sample size||CON%||LAB%||LIB%||UKIP%||Green%||Error%|
|TNS BMRB||30 Apr 2015 - 04 May 2015||1,185||33||32||8||14||6||9.0|
|The Sun/YouGov||04 May 2015 - 05 May 2015||2,148||34||34||9||12||5||9.6|
|Opinium||04 May 2015 - 05 May 2015||2,960||35||34||8||12||6||8.8|
|The Guardian/ICM||03 May 2015 - 06 May 2015||2,023||35||35||9||11||4||9.6|
|Daily Mirror/Survation||04 May 2015 - 06 May 2015||4,088||31||31||10||16||5||13.2|
|05 May 2015 - 06 May 2015||1,007||35||34||9||12||4||7.6|
|Evening Standard/Ipsos-MORI||05 May 2015 - 06 May 2015||1,186||36||35||8||11||5||8.8|
|Populus||05 May 2015 - 07 May 2015||3,917||34||34||9||13||5||8.8|
|Poll Average||30 Apr 2015 - 07 May 2015||18,514||33.7||33.4||8.9||13.0||5.1||8.5|
|SportingIndex||06 May 2015 - 06 May 2015||18,500||33.3||29.0||13.0||12.9||5.2||13.0|
|AVERAGE||30 Apr 2015 - 07 May 2015||37,014||33.5||31.2||11.0||13.0||5.1||8.6|
|Actual Result||07 May 2015||37.8||31.2||8.1||12.9||3.8||0.0|
None of the pollsters had a very good election. Every one had Conservative and Labour within one per cent of each other, compared with the actual gap of more than six per cent. At the time of writing, there is an investigation under way by the British Polling Council (the pollsters' trade body) into the polling errors. Amongst the pollsters, ComRes was the least inaccurate by a short head.
Electoral Calculus also used spread betting market prices from Sporting Index as well, because these had been successful in 2010. At this election, their performance was mixed. They were more accurate about the Conservative lead over Labour (seeing 4.3% instead of 6.6%), but they overestimated the Liberal Democrats. Their actual seat forecast was: Con 289, Lab 265, Lib 26, UKIP 3, Green 1, SNP 46. Their implied support figures were overall worse than the average of the pollsters, mostly because of their inaccurate belief in Lib Dem strength (or incumbency). Including them in the overall average made little difference in terms of vote share accuracy, but it helped in terms of seats.
Given the large polling errors, it is hard to tell how well the actual model performed. The model, which converts national support figures into seats, is only as good as its inputs. If the input polling data is bad, then the model output will be bad. This effect is sometimes described as "garbage in, garbage out".
But we can adjust for this by feeding the actual 2015 support levels, rather than the polling figures, into the model. Then we get the following result:
This is a much more accurate result, which helps confirm the fact that the prediction was wrong primarily because the pre-election polls were wrong. The prediction is still not quite exact. The Conservatives are a little low, and Labour are a little high, so that the Conservatives are shown just short of a majority, rather than just over. But it's not too bad, and the other parties are also predicted relatively well.
In terms of individual seats, only 36 seats are wrongly predicted, which is a good result. This compares well with 2010 when the equivalent figure was 63 mis-predicted seats, and 2005 which had 45 seats.
|Num||Seat Name||GE2010||Prediction||GE2015||County (Area)||Comment|
|1||Glasgow North East||LAB-54||LAB-10||NAT-24||Glasgow area (Scotland)||SNP strength|
|2||Coatbridge, Chryston and Bellshill||LAB-50||LAB-06||NAT-23||Glasgow area (Scotland)||SNP strength|
|3||Kirkcaldy and Cowdenbeath||LAB-50||LAB-05||NAT-19||Fife (Scotland)||SNP strength|
|4||Bristol West||LIB-21||LIB-02||LAB-09||Bristol area (South West)||SWest Lib weakness|
|5||Gower||LAB-06||LAB-09||CON-00||West Glamorgan (Wales)||Wales Con strength|
|6||Vale of Clwyd||LAB-07||LAB-08||CON-01||Clwyd (Wales)||Wales Con strength|
|7||Plymouth Moor View||LAB-04||LAB-05||CON-02||Devon (South West)||Con strength|
|8||Telford||LAB-02||LAB-04||CON-02||Shropshire (West Midlands)||Con strength|
|9||Derby North||LAB-01||LAB-04||CON-00||Derbyshire (East Midlands)||Con strength|
|10||Morley and Outwood||LAB-02||LAB-04||CON-01||West Yorkshire (Yorks/Humber)||Con strength|
|11||Southampton Itchen||LAB-00||LAB-02||CON-05||Hampshire (South East)||Con strength|
|12||Bolton West||LAB-00||LAB-02||CON-02||Western Manchester (North West)||Con strength|
|13||Cardiff North||CON-00||LAB-01||CON-04||South Glamorgan (Wales)||Wales Con strength|
|14||Warwickshire North||CON-00||LAB-01||CON-06||Warwickshire (East Midlands)||Con strength|
|15||Sherwood||CON-00||LAB-01||CON-09||Nottinghamshire (East Midlands)||Con strength|
|16||Stockton South||CON-01||LAB-01||CON-10||Teesside (The North)||Con strength|
|17||Broxtowe||CON-01||LAB-01||CON-08||Nottinghamshire (East Midlands)||Con strength|
|20||Amber Valley||CON-01||LAB-00||CON-09||Derbyshire (East Midlands)||Con strength|
|21||Wolverhampton South West||CON-02||CON-00||LAB-02||Black Country (West Midlands)||Marginal|
|22||Dewsbury||CON-03||CON-01||LAB-03||West Yorkshire (Yorks/Humber)||Unsplit anti-Con vote|
|23||Brentford and Isleworth||CON-04||CON-01||LAB-01||Hounslow (London)||Marginal|
|24||Hove||CON-04||CON-02||LAB-02||East Sussex (South East)||Anti-Con tactical|
|25||Enfield North||CON-04||CON-03||LAB-02||Enfield (London)||London Lab strength|
|26||Edinburgh South||LAB-01||NAT-24||LAB-05||Edinburgh area (Scotland)||Anti-SNP Tactical|
|27||Chester, City of||CON-06||CON-04||LAB-00||Cheshire (West Midlands)||Marginal|
|28||Wirral West||CON-06||CON-04||LAB-01||Merseyside (North West)||NWest Con weakness|
|29||Ealing Central and Acton||CON-08||CON-06||LAB-01||Ealing (London)||London Lab strength|
|30||Ilford North||CON-11||CON-11||LAB-01||Redbridge (London)||London Lab strength|
|31||Bath||LIB-25||LIB-08||CON-08||Bristol area (South West)||SWest Lib weakness|
|32||Dumfriesshire, Clydesdale and Tweeddale||CON-09||NAT-11||CON-02||Dumfries and Galloway (Scotland)||Anti-SNP Tactical|
|33||Yeovil||LIB-23||LIB-05||CON-09||Somerset (South West)||SWest Lib weakness|
|34||Twickenham||LIB-20||LIB-00||CON-03||Richmond Upon Thames (London)||Marginal|
|35||Southport||LIB-14||CON-06||LIB-03||Merseyside (North West)||NWest Con weakness|
|36||Carshalton and Wallington||LIB-11||CON-09||LIB-03||Sutton (London)|
[Note the use of Slide-O-Meter notation of "CON-03" to mean a Conservative majority of 3% which is used in this table. Majorities are rounded to the nearest integer percentage, so "CON-00" means a majority of less than 0.5%.]
There are a number of stories here. In outline they are: