On the predictive properties of the Iowa & NH primaries, redux

Are there any swing states that are small enough for retail campaigning to be vital yet have substantial minority populations?

That is a good point though. If IA and NH are tests of electability, they function better than most states. There are two paths to electability in this day and age: win more of the white vote than average, or increase minority turnout above average. It seems to me that the latter question is adequately answered for Democrats in South Carolina as it is. The former is answered in IA and NH. Obama proved both in 2008 by winning South Carolina with high turnout, and winning IA(which also had record turnout despite being white).

For the purposes of the predictive power of these two primaries compared to previous years in picking the eventual nominee the percentage win should be taken into account.

So far we have
Iowa: Slight win to Clinton by 0.25 percent
NH: Landslide to Sanders by 20 percent

[Minor hijack]

This quote from a year ago is interesting. I think it properly explains what happened in 2004 and why.

Today in 2016 we have an R base utterly consumed by hate for Obama (who isn’t running this time, despite how some folks seem to be acting).

The interesting question is what impact this will have on the R’s choice of nominee and what that’ll cause in November.
[/mh]

To relate the above to the rest of this thread I think we’ll see the IANH predictive model hold up for the Rs in 2016. The two mostly-hate-based candidates took a first place apiece, and we’re on-track for one or the other to be the nominee.

Then we’ll get to see whether adaher’s rule applies.
There’s an adage in sports that defense wins championship games. It’s certainly widely accepted as true in NFL & MLB. I hold that it applies to NHL also, but I admit I don’t know the NHL experts’ opinions.

I submit the corresponding adage in politics is “Anger-driven candidates lose the general.” All emotion is blinding, and spite / anger most of all. I just hope it holds this time despite the propaganda machine’s noisy attempts to make anger a mainstream value.

If Trump or Cruz are the nominee, the chance of a loss is definitely higher. Although Cruz is hard to figure. He’s so smart you wonder if he hasn’t gamed all this out and already got everything in the bag.

You are welcome to do that! Feel free to go back through the results from 1976 to the present, and post on the extent to which the percentage of the wins has made a difference. Seriously, knock yourself out - I’d welcome such a post here.

But if you’re going to say that percentages should matter this time, that should be based on some data showing that that’s mattered in previous cycles.

I’ll give you one counterexample: 1988 on the GOP side. Dole beat Bush Sr. in Iowa, 37-19, then Bush beat Dole in NH, 38-29. As we know, Bush won the nomination, even though Dole won Iowa by a much bigger margin than Bush won NH by.

Maybe that was an outlier. But you’ve got to make the case. You can’t just wave your hands and say that margin of victory has predictive power, on the basis of your gut feeling.

Well, you can, but in the words of Richard M. Nixon, “that would be wrong.”

I was just thinking along these lines myself. How do the age demographics of those states line up with the national statistics?