Making sense of the polls

Making sense of the polls

Clinton clearly ahead of Trump in US Presidential contest

I have not written about the US Presidential election since late June, owing to coverage of the Australian Federal election. The US election will be held on 8 November.

In July, following the FBI’s damaging report on Clinton’s emails and international terrorist attacks, Trump reduced Clinton’s lead from about six points to three. Trump took the lead briefly after the Republican convention on 18-21 July, but the Democratic convention on 25-28 July pushed Clinton back in front by about eight points, and she has held this lead since, partly owing to Trump’s many controversial statements.

Clinton currently leads Trump nationally by 43-37 when third party candidates are included, or by 47-40 head to head. Trump’s favourability is very poor at 63-35 unfavourable, while Clinton’s is better, but not great, at 56-42 unfavourable. By contrast, Obama has a 50-46 job approval rating.

I have written previously that the popular vote is not decisive; the electoral college is. Current polling has Clinton leading in states worth 343 electoral votes (EVs), to Trump leads in states worth 195 EVs. Clinton has double digit leads in states worth 231`EVs and leads of 5-9 points in states worth another 89 EVs. 270 EVs are required to win the Presidency, so Clinton is in a strong position.

In the Republican primaries, Trump had a core support of about 35%, and was able to win the nomination because the rest of the vote was split between several candidates for a long time. In the general election, there are only two major party candidates.

Had Clinton been facing a traditional Republican, she could be losing now; her favourability ratings are poor, and retention of the White House for three successive terms by one party is difficult. However, university-educated whites and Hispanics are much less likely to vote for Trump than they were for Romney in 2012, and Clinton should win easily.

UK Brexit political fallout

On 23 June, the UK voted to leave the European Union by a 51.9-48.1 margin. The day following Brexit, Conservative PM David Cameron announced that he would resign once a new leader was chosen. The Conservative party MPs were to whittle the candidates down to two, with a membership vote to decide which of these two would be the new Conservative leader, and thus UK PM.

Home Secretary Theresa May won a large majority of Conservative MPs on 7 July, with Andrea Leadsom a very distant second, ahead of Michael Gove. Leadsom then withdrew from the membership vote on 11 July, allowing May to become PM on 13 July, much sooner than was expected.

On the Labour side, there has been a push from the parliamentary party to remove Jeremy Corbyn as their leader. Corbyn is an old-style socialist, who is, in my opinion, completely unelectable at a general election. He won the Labour leadership in September 2015 because Labour’s membership has become unrepresentative of Labour voters generally, let alone the general electorate.

At the UK 2015 election, Labour earned 9.35 million votes. At the leadership election, Corbyn won with about 250,000 votes out of 422,000 - just 4.5% of Labour’s general election vote.

Corbyn was somewhat blamed for Brexit because of his muted opposition. After Brexit, most of Labour’s shadow Cabinet resigned, and over 80% of Labour MPs voted no-confidence in Corbyn on 28 June. However, Corbyn has vowed to remain leader unless the members vote him out.

Angela Eagle announced that she would challenge Corbyn on 11 July, and Owen Smith announced his challenge on 13 July. Eagle withdrew on 19 July after it became clear that Smith had more support from the parliamentary Labour party.

The leadership election will take place by mail, with a result to be announced on 24 September. The most recent poll of Labour members, a YouGov poll of 1030 respondents taken 15-18 July, has Corbyn leading Smith 56-34.

Corbyn has very low ratings from the general public, and his likely re-election as leader will very probably be a disaster for Labour at the next general election. Polling currently gives the Conservatives a 9-point lead, though some of this lead is due to May’s honeymoon.

Australian Senate: Roberts’ 77 below the line votes not a reason to doubt new system

Much has been made of how One Nation’s No. 2 Queensland candidate, extreme climate change denier Malcolm Roberts, was elected despite winning only 77 below the line votes. The truth is that in most states, below the line votes have very little impact. Many major party candidates won with much less than 0.1% in below the line votes.

Roberts won because he received 0.19 quotas after Hanson was elected on her surplus. He was then able to benefit from above the line preferences that came from many other populist right wing parties, and this allowed him to pass other candidates who seemed in a better position on primary votes. Roberts finished with 0.78 quotas, well ahead of Family First with 0.69.

According to Kevin Bonham, the below-the-line voting rate was up at this election in all states. The biggest increase was in Tasmania, from 10% to 28%. In WA, Queensland, Victoria and NSW, the below the line rate was 5-6% compared with 2-4% in 2013. This can be explained by ease of voting: only 12 preferences were required at this election, rather than for all below the line squares to have a preference.

While the Senate informal rate increased 1.0 points from 2013, the informal rate in 2016 was almost the same as in 2001, 2004 and 2010. The new system had little impact on informality.

NT election on Saturday

The Northern Territory election will be held this Saturday 27 August. Polls close at 6:30pm Melbourne time. I have not seen any NT polls since a late July one, which had Labor ahead by 64-36. The NT legislature has 25 members, elected by single member electorates using optional preferential voting.

I will be writing for Election Watch on the US election from early September

I have accepted paid work for US Election Watch. For the next couple of months I will be a regular columnist on US politics for the University of Melbourne’s Election Watch. Election Watch will begin publishing on the US Presidential election in early September. I will still be able to cover non-US politics for The Conversation.

On Saturday 6 August, I officially received my PhD for my thesis, titled: “Transforms and Truncations of Time Series”. I am now an Honorary Associate with the School of Mathematics and Statistics.