ABCDEFGHIJKLMNOPQRSTU
1
MethodAuthorForecast Con/Lab LEAD 2015last updatedNotesLinks
2
3
By-election SwingbackRodCrosby-0.46%21/11/2014Performed well up to 2010, when the Tories did significantly better than predicted. Analysis suggested this was due to success in the TV debates, introduced for the first time, and were worth about a 1% swing to the Tories (20 seats)
4
5
6
7
Local ElectionsChris Prosser5.00%06/06/2014regression of local NEVs compared to General Election outcomeshttp://politicsinspires.org/four-locals-european-2011-14-elections-might-tell-us-2015-general-election/
8
http://www.academia.edu/1530490/Do_local_elections_predict_the_outcome_of_the_next_general_election_in_the_UK
9
http://blogs.channel4.com/factcheck/factcheck-farage-westminster/18277
10
11
Local ElectionsRodCrosby8.38%25/05/2014simple regression of the average Opposition local election NEV lead versus subsequent general election lead (1979-2010). Forecast based on 2011-14 Local Elections Rallings & Thrasher National Equivalent Voteshares.https://docs.google.com/spreadsheet/ccc?key=0AswNZWYSW1uvdHJ1OGRHZXQ5Y2lzYVVEV1VJc190QWc#gid=0
12
http://www2.politicalbetting.com/index.php/archives/2014/05/27/guest-slot-rod-crosby-the-bell-tolls-for-labour-and-miliband/
13
14
15
Opinion PollsStephen Fisher2.00%15/04/2015The approach is broadly to predict the next election based on current opinion polls and the track record of polls in previous electoral cycles, allowing for change in opinion in the run up to the election.http://electionsetc.com/
16
17
18
19
Opinion Polls2009-10 poll swingback-1.80%12/03/2015Models based on the swingback last time, based on estimate of current polls, 2009-10 Kalman filtered polls and 2010 election result
20
21
22
23
Election forecastHanretty et al1.80%16/04/2015?http://electionforecast.co.uk/
24
25
26
27
PM ApprovalLebo & Norpoth8.32%16/04/2015Performed excellently in 2010, forecasting 311 Tory seats, 265 Labour. Highly dependent on the accuracy of final PM approval polls. Forecast based on April 2015 IPSOS-MORI PM approval/two-party votes.http://ms.cc.sunysb.edu/~mlebo/Lebo%20and%20Norpoth.2012.pdf
28
29
74 seat lead
30
31
32
Average3.32%16/04/2015May as well throw in an average prediction, as they are quite widely spread at the moment.
33
34
35
36
SOME CAVEATS: all of the methodologies are obviously based on data collected from a period of single-party government and a broadly 2/3 party system. The political landscape since 2010 now looks rather different. Who knows if the models still apply, or to what degree?
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100