Posts Tagged ‘political predictions’

youcan

 

We haven’t been doing very well with our predicting, recently. There: that’s a sentence you don’t normally expect in the world of political punditry.

But the fact is we got the Brexit vote totally wrong – we were hardly Robinson Crusoe in that regard, but that doesn’t matter – and we also failed to predict a Conservative overall majority in the UK, but again, almost no-one got that right either and it was a very close run thing … from a psephological point of view the change in the Lib Dem seats in the West can literally be assumed to have taken place in the last 48 hours, so we weren’t the only people surprised.

But for a year now, we have been confidently predicting a Clinton victory in the 2016 Presidential election. Our analysis has ranged from “massive Clinton victory” to “comfortable Clinton victory” and at no stage have we EVER speculated that Trump could win.

And today, despite all the recent froth and bubble in the media, we repeat our prediction. Clinton’s got this. With a margin somewhere between “just” and “just about comfortable”. Which considering she has been subject to the longest period of personal vilification ever seen in modern political history, much of which has been entirely politically motivated and rank untruths, is really quite a remarkable thing. She has been pilloried, lied about, accused of ludicrous things, with a level of pure vitriolic hatred that can only be described as anti-democratic and neo-fascist in its expression. And she’s a woman. And anyone who thinks that there isn’t evidence of blatant sexism in some people’s denial of her right to be President is simply naive, just as many of the attacks on her health and stamina – which seem to be just fine when you consider her punishing schedule – are codes for “weak, poor little woman thing”.

Let’s be clear: we have previously pointed out at length that “No”, Clinton is not a perfect person. Who is, after all? Let alone in the field of politics. She is, however, a shining knight clothed in holy armour wielding a blazing sword of righteousness compared to Trump. That a large percentage of intending American voters can’t see that is truly distressing and disturbing to the rest of us.

We can’t bring myself to be angry with every Trump voter. After all, they’re perfectly entitled to disagree with us, and we know some of them personally, and they are patently obviously decent folk. Social media has dominated where they are getting their information from, and this is what’s dominating their social media , a highly sophisticated misinformation campaign mounted deliberately, and with malice aforethought.

So what we have been witnessing is a perversion of democracy. Let us hope that enough Americans agree so that after Tuesday we see some real and meaningful reform to both media coverage and the tone of campaigning. This morning on radio in Melbourne one of the more senior (and least alarmist) hosts remarked “America is broken. It’s like Humpty Dumpty, and I cant see how it’s ever going to be put back together again.” Which is a miserable thought for all who treasure much of what is good – even great – about America.

Anyway, back to the impending result. A lot of people have been opining that some Trump voters are hiding their intentions from pollsters because they’re embarrassed to admit they support him, and arguing that the phenomenon is “like Brexit”, which is all very well except right up to the Brexit vote it was always going to be close (Brexiteer Nigel Farage famously remarked that a 52-48 vote in favour of Remain should be ignored, then promptly changed his tune when his side won by that very margin), and speaking out against the EU was never something that the vast majority of Britons had any problem with.

It should also be noted that for someone to be so embarrassed to be supporting Trump that they would have to lie to pollsters yet still be intending to vote for him would require some very complex mental circumlocutions. And given the hurricane of criticism she has received, we think it just as likely that some Clinton supporters might not wish to express their support for her, so any “ashamed of Trump” effect would in all probably be evened out by an “ashamed of Clinton” effect. Indeed, if anything, the “enthusiasm” levels that are measured are lower for Clinton than Trump, so that effect would be greater.

So we simply do not believe, as the Trump side has been furiously talking up, that there is some vast unspoken well of support for the Republicans.

Essentially, we think the opinion polls have it about right, which is somewhere in the region of a 2-3% lead in the popular vote for Clinton, which is now climbing again, but a stronger lead for her in some key battleground seats. We think Trump may actually do better in some of the solid “red” seats than is currently expected, except it won’t matter because he’ll be piling up votes where he doesn’t need them.

There has been much talk about which the key swing states are. We all know about Florida, where we suspect Trump may just pull it off, following an erosion in support for Libertarian Gary Johnson which has caused a drift back to him, as people focus on the main game. But Florida really is too complex to call with any degree of certainty. Two other factors are complicating matters. Firstly there is some strong evidence that a significant percentage of early Republican voters have backed Clinton, which in a state with a very significant early voting percentage, and where the Democrats ground game is markedly better, may just tip the state back to her. It was won by Obama, remember. Secondly, Latinos in the state are voting in increased numbers, and as we have opined previously, Trump has very good reason to fear an uptick in the Latino vote. Because of our political preferences we are hopeful of a Clinton victory, but the psephologist in us urges caution. And in any very close election, we are minded to remember Bush v Gore, which we remain convinced was nothing more nor less than a judicial coup. Now, though, the Supreme Court is split 4-4 between Liberals and Conservatives, so any similar farrago this time may be avoidable … Hillary Clinton is ahead 48 – 42 percent among Florida voters who already have cast ballots.

Let us hope this election doesn’t come down to lawyers at dawn. One calculation has a 17% chance that Florida will be the “tipping point” state, yet again.

FLORIDA VERDICT Too close to call/Very close Clinton victory

Moving up the country, the next vital state is North Carolina. There were some early indications that Clinton might be in trouble in a state with a large black vote where there was less enthusiasm in black voters than in the Obama elections, in a state won by the Republicans last time, and in areas where Republican voter suppression has been seen at its most naked. The voter suppression laws have been largely declared invalid, but possibly too late to rescue the situation. Then again, President Obama, who is still a talisman for the African-American vote, has been strategically deployed to “get the black vote out”. If he succeeds, what might have been the narrowest win for Trump may turn into a narrow win for Clinton. There’s no question that the race is tight, but we perceive Clinton inching ahead in the last couple of days. In particular, a normally very reliable poll now has her up by three having previously had Trump ahead by four just a couple of weeks ago. In a reliable poll using the same methodology, that’s a significant movement, and in the last week Clinton has seemingly risen slightly further. The only impossible to discern factor is how far Libertarian candidate Johnson will fall. Five Thirty Eight actually says Trump is more likely to win, but we think there may be enough African-American enthusiasm to carry the day for the Democrats.

NORTH CAROLINA Too close to call/Very close Clinton victory

The next crucial contest is Ohio. We think that this can now be safely called in the Trump camp. He has been improving there steadily, and there is, clearly, a generalised move to the Republicans in the central states. Some of the key factors here are that Clinton is doing significantly less well with female voters here than generally, and Trump has garnered an historically high number of male voters.There is also a strong sense amongst the working class and non-college educated vote that they are being ignored by the elite, especially as regards trade deals and employment initiatives. The notably right wing Governor, Kasich, has ironically and pointedly refused to back Trump, but his general stance seems to have solidified Republican support.

OHIO Trump

Moving to the right, but only geographically, Pennsylvania seems equally locked in the Clinton column. The state seems to be delivering a solid lead of 4-6% for Clinton in survey after survey no matter how you dice and dust the results. That’s now too big for Trump to overhaul except in some mythical scenario where his vote is being under-estimated by 10-20%, which isn’t going to happen. A potential transit strike in Philly won’t help poorer voters of either party to get around on polling day. It remains to be seen whether the City’s attempt to injunct the strike is successful.

PENNSYVANIA Clinton

It is worth pointing out that no one has been elected president since 1960 without carrying two of the three key swing states, Florida, Ohio and Pennsylvania. This year, North Carolina has been added to that mix. We see the four states leaning to Clinton overall – just. Are we confident in that prediction? Not especially. In Florida and NC we could be dead wrong, as the current polls are well within margins of error. Interestingly, though, that can cut both ways. Take 3-4% off Clinton and Trump does very well indeed. Add 3-4% to Clinton and it becomes a Clinton landslide.

Can we get a better idea by looking at other close states? Maybe. But first a word overall. Nate Silver’s normally highly reliable 538 site has a movement back to Clinton from Trump in terms of their chances of winning of what may be a hugely significant 0.8% from the fourth to the fifth of November. This presumably reflects both the FBI’s confirmation that Clinton is to be charged with nothing regarding the endless email scandal, and the following:

The small state of Nevada should deliver it’s 6 votes to the Republicans without any difficulty. Except … the Democrats ground game here has been especially aggressive, particularly via organised labor and targeting the Latino community. If Trump loses Nevada – and he could – he’s in trouble.

Whilst New Hampshire has come back towards Trump in the very recent past, Clinton is still solidly ahead in the vast majority of polls, although there have been a couple that have it line ball or leaning Trump, but these come from polls that seem less reliable than some others, and one which is positively partisan. The role of the Libertarian candidate here will be crucial in a state where “alternative” candidates often do well. How well his vote holds up will probably decide the state, and at the moment he seems to be hanging onto enough votes to cruel things for Trump. It’s only 4 electoral college votes, but they’re four votes who if they went to Trump it would suggest a real Republican surge.

In Michigan, with its important 16 electoral votes, Clinton is holding onto about a 4% lead, with Gary Johnson’s vote holding much steadier than in some other places. Whilst the race unquestionably tightened in the last half of October the situation now seems more stable for Clinton, and possibly moving back to her slightly. The opposite is true in Iowa, where Trump is strengthening his position, as with many mid-West states.

Virginia was close for Obama but it is becoming increasingly suburban and – as it does – liberal. If Clinton were to lose it – we don’t think she will – she can kiss her hopes goodbye.

So: our prediction?

We predict that without Pennsylvania, Trump cannot win, even if he carries Ohio and Florida – unless he is able to also capture Michigan and Wisconsin. But both states have voted Democratic in the last six presidential elections and Clinton is ahead in both.

One of the interesting factors will be how early Florida declares – or is declared – by the media. Beyond the very fact of the significance of its votes going to Trump, because of the time difference a very early declaration for Trump could cause a cascade effect through the centre and west.

A Trump win in Pennsylvania would be very difficult for Clinton to make up. The loss of Pennsylvania together with Florida would be a real blow to her chances.

We see Clinton with anything from the barest 270 electoral college votes up to about 290-300. Trump probably about 230-256, maybe a few less.

Which means Trump will do better than anyone ever imagined possible until recently. But no: he isn’t going to win.

Betting odds are currently 4-1 or so ON Clinton and 4-1 or 5-1 AGAINST Trump. That looks about right.

Well, for today, at least. We’ll see more polls and analysis tomorrow, but till then, well, that’s what we think.

UPDATE as at 14.25 AEST

Polls have firmed for Clinton overnight with it now looking likely she will take Florida, unlikely she will take North Carolina but not impossible, not going to take Ohio, will take Pennsylvania, and will take Michigan and Nevada.

Somewhere, a fat lady is singing. Clinton wins.