graphic from BBC website here
I was right that Labour would greatly increase it’s majority in the Oldham East and Saddleworth by election (which was the easy part to predict), but wildly wrong in predicting that the Lib Dem vote would collapse, with the party losing votes to both Labour and the Conservatives.
Instead, while a few thousand Lib Dem voters many Conservatives seem to have voted tactically for the Lib Dems, as the Conservatives had little chance of winning the seat. This should have been predictable given that Labour voters have often voted tactically for the Lib Dems in constituencies that Labour had little chance of winning – and since the opposite happened in the May 2010 General Election.
This is all a bit uncertain as you can’t necessarily compare by-election results with General Election results, because the turnout in by-elections is always lower (in this case 48% of registered voters voting compared to 61% in the invalidated 2010 general election result in the constituency).
However it’s hard to see a better explanation for the Conservative vote being less than half what it was in 2010, Labour’s total number of votes increasing slightly on a much lower turnout; and the Lib Dems not only maintaining their share of the vote but increasing it slightly despite a sharp drop in their poll ratings over their participation in the Coalition and Clegg’s broken campaign pledge on tuition fees.
I don’t feel too bad about getting it partly wrong, since the only solid rule in predicting what decisions large numbers of people will make and why in the future, often after unpredictable events, is that you can’t
In the 2010 election the Conservatives overtook the Lib Dems in votes in many seats that they’d been behind them in in 2005 as Conservative voters, seeing their party had a chance nationally, seem to have switched from their previous tactical votes for the Lib Dems. (That’s the best explanation I can see for the rise in Conservative votes and the fall in Lib Dem ones – e.g Lanark and Hamilton East in 2005 and in 2010).
If it’s the case it also suggests that some voters don’t make rational decisions but are happy to cast a “wasted” vote so long as they’re voting for what they think might be the winning party nationally - even if they’re voting in ‘safe seats’ where, under the backwards first past the post election system, their votes count for nothing anyway, unless they voted for the winning candidate in that seat.