Norpoth Statistical Prediction of Trump Win

[Michael Stuart Kelly]

Norpoth Statistical Prediction of Trump Win

Does anybody know anything about this guy Helmut Norpoth? He's reputed to have one hell of a degree of accuracy in his predictions.

He claims 87-99% chance Trump will win the popular vote in the election in November. See his site:

Primary Model

This is more for the polling/statistics wonks. I'm more into the storytelling, narrative and persuasion techniques.

For a quick look at Norpoth, Lou Dobbs interviewed him recently:

Also, just in case that video gets taken down, Alex Jones commented on this interview.

I'm pretty comfortable with this professor, not just because I support Trump. It's because the media is so over-the-top these days against Trump, it's obviously manipulated and does not represent the audience. Just like, I believe, current polls do not represent the public at large.

Emotionally, though, Norpoth's prediction does feel good. And it would feel good even if it were shaky.

 

Fortunately for Trump supporters like me, this looks like it is a serious model.

Michael

[william.scherk]

Norpoth issues his forecasts fairly early in the races, and the last three predictions are available on his site.  For 2016, it is pretty basic.   Norpoth takes in the winners of the first primary/ies.  He adjusts for incumbency and for a 'pendulum' effect or pattern (where it is much more difficult to have a 'third term,' ie, an incumbent party winning three straight presidential elections).

So, in 2008, Norpoth predicted  "Democrat Hillary Clinton would narrowly defeat Republican John McCain in the November general election (50.5 to 49.5 percent of the two-party vote). [...] Finally, in match-ups between primary losers, both Obama and Edwards would beat any of the Republicans, and quite handily so in most cases."

For 2012, Norpoth predicted "Barack Obama will win a second term in the November election, defeating Mitt Romney or any other Republican challenger by a comfortable margin."

This year, Norpoth is firm: 

 The Primary Model predicted on March 7, 2016 that Trump would defeat Hillary Clinton with 87 percent certainty.

[...]

The forecast of a near-certain Trump victory at a moment when he was trailing both Clinton and Bernie Sanders in every poll, some by double-digits, was greeted with a heavy mix of shock, cheers, amazement, and derision, much of it on social media but also regular media outlets.  Many offered bets against the forecast, gleeful that it would turn out wrong. There is nothing to add to or subtract from the March forecast here. It was unconditional, final, and not subject to updating.

So, there you go. 

I think the model has face validity. In this year, as all years, nothing extra has been added to the mix of data entering the model. This is an election like any other in terms of input data from primaries, and the modelled effects of incumbency and the Pendulum.   In his words, "After two terms of Democrat Barack Obama in the White the electoral pendulum is poised to swing back to the Republicans in 2016."

Now, could he plausibly be wrong about the pendulum?  If this firm forecast turns up turnips, he may be able to explain it by virtue of a Factor X (idiosyncratic effects particular to the pertinent race, extra 'qualities' in play).  The only factor I would put a name on is probably Trump.  The Trump X factor.   The outlier, the exception to the rules. 

Could his model plausibly fail to account for an unique X factor? Or does the pendulum factor override it?  The clarity of the model is a nice quenching drink for a Red Hat, but might not satisfy the wonks in blue.  If I were a Red Hat, I would be happy with this, as I would be happy clutching straws under other circumstances. That the polls being 'off' is a kind of filter one can apply to any assessment of weakness or challenge to the Trump enterprise. The Norpoth model does not look at anything beyond the first primaries. It has been uncoupled from events since March 7.

I suggest noting this forecast against forecasts that have evolved over the campaign. I advise focussing the campaign mind on the foremost challenges, and giving (for arguments' sake) the benefit of the doubt to Pennsylvania, Ohio, Florida polling, assuming-so-not-to-be-complacent that the numbers are in the ballpark.

In which case, singing La La La It's all Phony at 'danger' signals for the campaign can be less than efficient in getting out the vote.  Knowing where there are weaknesses is campaign intelligence. If you don't believe you have to mount an effort to change the signals, then you are going to be a loser in some measure. 

That said, whining about losing only to cheating and rigging is a Factor Z I haven't incorporated in my mental models of Who's On First. Everything is a red rosy glow of Trumpism depending on where you stand, I think, if in the middle of a movement.  To maintain the bubble against X factor, I'd have to  de-legitimize all corrections, cautions and warnings, except as per Z factor.  It is hard to keep both X and Z out of the larger discussion, except by shutting a person's argument out of some kinds of conversation.

(I haven't done any digging into any but Norpoth's own site yet)

Edited August 18, 2016 by william.scherk
Pith, poth not path.

[Michael Stuart Kelly]

1 hour ago, william.scherk said:

That said, whining about losing only to cheating and rigging is a Factor Z I haven't incorporated in my mental models of Who's On First.

William,

The way you frame this is not the way Trump supporters see it, nor the way I believe Trump sees it. I don't seriously contemplate a Trump loss. I don't think he does, either. (Caveat: 99.99% sure of a victory just to cover the anomaly.  )

The fact is that I, and I am certain Trump, believes the Democrats will cheat as much as they can get away with. They proved it in the primaries. So this is more of a rallying cry to get people to help watch the voting sites and the election process in general than anything else. It's a form of demonizing dirty rotten scoundrels to mobilize a group to action than preemptive excuse-making.

The Democrat side talks about whining and blah blah blah, but I don't see acceptance of defeat in the charisma of Trump supporters. Not at this point in time. Instead, I see an upbeat belligerence, seriousness and swagger.

Check this lady out, for instance. I don't know her. I merely picked her at random because this video appeared on my Facebook feed, shared by a Trump supporter I don't know, but who friended me due to a comment I made. There are people like this all over Trump-land.

Michael

 

[william.scherk]

 

1 hour ago, Michael Stuart Kelly said:

2 hours ago, william.scherk said:

I suggest noting this forecast against forecasts that have evolved over the campaign. I advise focussing the campaign mind on the foremost challenges, and giving (for arguments' sake) the benefit of the doubt to Pennsylvania, Ohio, Florida polling, assuming-so-not-to-be-complacent that the numbers are in the ballpark.

In which case, singing La La La It's all Phony at 'danger' signals for the campaign can be less than efficient in getting out the vote.  Knowing where there are weaknesses is campaign intelligence. If you don't believe you have to mount an effort to change the signals, then you are going to be a loser in some measure. 

That said, whining about losing only to cheating and rigging is a Factor Z I haven't incorporated in my mental models of Who's On First. Everything is a red rosy glow of Trumpism depending on where you stand, I think, if in the middle of a movement.  To maintain the bubble against X factor, I'd have to  de-legitimize all corrections, cautions and warnings, except as per Z factor.  It is hard to keep both X and Z out of the larger discussion, except by shutting a person's argument out of some kinds of conversation.

I don't seriously contemplate a Trump loss.

If you don't seriously contemplate a Trump loss in the following three states, then you may be making a cognitive error.  I mean in the sense of accurately assessing "Who's On First" and directing resources to Get Out The [Trump] Vote.

The red rosy glow of Trumpism refers to a kind of goggles worn. They polarize out any 'danger' signs in the polls for the trio of states by assuming the polls are rigged.  It becomes concurrent with Mr Trump's declaration that the only way he will plausibly lose Pennsylvania is by some unspecified blotch of cheating.  

Now, I understand your identification with Trump's day-to-day thinking. Or your guesswork at what he is thinking about 'danger signs,' or your evidence that other people are Trump supporters.   

But back to your model. The polls are wrong, so they don't factor.  The cheating can't be specified, so it will be a hella job to root it out this year, The cheating only factors on a win by Clinton, otherwise the polls on Election Day are fair.

How can we accurately and reliably estimate 'public opinion' or intentions to vote?  In your model, there is a large but unspecific and undetected 'extra' bloc of voters that the polls are either missing -- who will rise up to get registered and flow into the precincts.   And this flow will overcome the degree (unspecified) of cheating otherwise engaged in. And somehow a contingent of unaccredited yahoos at polling precincts not their own are going to keep the bad thumb off the scales. Or something.

Anyway, my main point was that ignoring 'danger' signs is not good politics.   I can't estimate Factor Z (cheating) but I can be aware of its appearance in the Trump camp's boogeymen.  

So, the depressing polls aren't real boogeymen, but the (murkity murk murk) is.   This makes me think of a very sad fudge cake.   The baker blamed it on the fudge.

Contemplate a Trump loss in, say, Florida.  What happened?  Wuz he Rawbed?  Is that the way we must spin it, given our vantage today?

PS. -- bonus question. Does anyone endorse that Facebook video?

Edited August 19, 2016 by william.scherk

[Michael Stuart Kelly]

1 hour ago, william.scherk said:

The cheating can't be specified...

William,

Where on earth did you get that idea? The volunteers are going to get orientation about what to look for. (Voting machines, dead people, out of town folks voting often, etc.) I expect there to be a lot of data sharing through smartphones and tablets this go around. Pictures, etc.

Here's a mini-prediction of mine. I predict a lot of shit will hit the fan around election time as one illegal immigrant after another gets exposed in the press by the citizen brigade because he or she voted. And the proof will come in all sorts of shapes and forms. I would not want to be an ICE agent at that time.  

Anyway, there's a point where we are talking past each other and it's in the way of thinking. The approach.

In your mind (going from your words), I am doing magical thinking or something like that when I don't contemplate a Trump loss. And that magical thinking is due to me ignoring indications from the establishment corporate media and establishment polling companies. In my mind, I am building a project. Or maybe playing a football game. I'm in that frame of mind.

So I don't worry about losing while I am doing it. I do my best, I keep doing it as hard as I can and I ignore the naysayers on a deep level. Victory is a project to be earned, not a gift to be bestowed by an elite.

Trump actually won the primaries that way. Shit, if he (or I) had listened to the people you want me to take seriously as they kept foretelling his doom, he would have dropped out long ago.

Here's a hint at the difference between mentalities. Narcissism is not confidence, yet Trump critics always peg him wrong and write him up as a narcissist. Likewise, focused effort and persistence during slumps is not magical thinking. Yet look how we talk past each other right now...

The Clinton machine took out Sanders. That is true. They had the elite tucked up their sleeve like a pair of aces. But they are not going to do the same with Trump because there are no big bosses in the background to corrupt like Sanders faced with the DNC. 

And there is a crapload of grassroots folks who are now engaged in keeping this contest honest. And that group will grow massively by the election time.

Michael