It is the holy grail of the presidential campaign: knowing which states will be decisive in the electoral college. We have our guesses: States like Arizona, Michigan, North Carolina and Wisconsin are widely expected be among those Top swing conditions for 2020. But how sure can we really be in these expectations five months before the election?
To find out, we went back and checked which states were election workers The cook’s political report From mid-June 2004, 2008, 2012 and 2016, these were swing states. We then compared these early ratings to the actual swing conditions after the elections. (To put it bluntly, Cook is doing a great job and we are not trying to select it. We are only using it as a proxy to see where conventional wisdom was at the time. It is actually surprisingly difficult to find historical sources for it and Cook is one of the few points of sale that we found consistently with early presidential race predictions from 2004. In an email, Charlie Cook, founder of The Cook Political Report, said he views his ratings as a rough guide, especially on this one Time in the cycle: “I was always of the opinion that we are not trying to predict results, but rather helping our readers to recognize which states are safe in the column of a party, which are not, but which could come into play, which are the ones competitive, but one side seems to have an advantage, and which ones are really, really close. “)
[Related: Our Pollster Ratings]
What is important is that much of the accuracy of these early assessments depends on how you define a swing condition. For example, “swing state” could simply mean a highly competitive state – a Throwing condition in Cook’s jargon. But it could also mean a Bellwether condition – Also known as a state, the results of which closely match the national referendum. That may sound like semantics, but it is actually an important distinction: with a roughly bound choice, throw states and Bellwether states are more or less the same. For example, in an election where a candidate leads by 10 points at the national level, the Bellwether states may not be very close. So it depends on how you define a swing state.
And as it turns out, early assessments are often wrong about which states it will be Toss-ups in November. But as we will explain, it does not mean that they are useless or even bad. In fact, Cook has a pretty good track record in identifying elections Bellwether states early, and that’s probably more important.
First, let’s look at which states Cook was expecting at this point in the cycle of the last four presidential elections, and which states have actually led to maladministration. The blue-green states in the table below are those that Cook correctly rated as an error. The pink ones are conditions that Cook classified as errors, but which have not turned out to be that close. and the gray ones are conditions that Cook didn’t judge to be errors, but landed close enough to count as one.
As you can see, Cook’s task invocation report is mixed. in the June 2008Ohio was the only state that it correctly called. in the End of May 2016It correctly identified New Hampshire and North Carolina as mistakes, but missed nine other states that turned out to be narrow. On the positive side, in April 2004Cook correctly predicted the status of seven states, missed only four, and misled only in Florida and Missouri. And in Late May 2012It has linked all four possible error states correctly. While it may appear at first glance that they have thrown too far a net by also classifying Colorado, Iowa, Nevada, New Hampshire and Pennsylvania as tossups, these states have hardly missed the cutoff for actual tossups (then President Barack Obama) won them 5 to 7 points each).
However, what went wrong in 2008 and 2016 was not that Cook misunderstood voters in certain states. It is not clear how competitive (or not competitive) the President’s race would be as a whole. (To be clear, this is not Cook’s fault. There was no way anyone could expect the economy to collapse in the fall of 2008 and Obama would lead to a landslide victory, or the letter from FBI director James Comey at the time October 28 to Congress, the 2016 campaign seemed to make a normal survey mistake of a Donald Trump victory.)
This is clear when you look at the states where Cook’s spring ratings implied that it was Bellwether: most of them turned out to be real Bellwether in November.
At the end of May 2016, Cook had wrongly believed that former Secretary of State Hillary Clinton was a general favorite, but she had rightly predicted that Colorado, Florida, Michigan, Nevada, Pennsylvania, Virginia, and Wisconsin would accurately reflect the national referendum (though three) other states too). Similarly, Cook’s ratings in June 2008 implied a presidential race that was too close to call it, but even after the race shifted in favor of the Democrats, four of the six states that Cook considered the most likely Bellwethers proved to be as Bellwethers (though it did) four other states missed Cook). And in 2004 and 2012 Cook had more success in pre-identifying Bellwether states than in pre-identifying toss-up states.
Where will we be in 2020? Well, as of June 16, The Cook Political Report lists six states as mistakes: Arizona, Florida, Michigan, North Carolina, Pennsylvania and Wisconsin. And since Democrats in states with only 232 votes have the advantage and Republicans in states with only 204 votes have the advantage, Cook also expects these six states to play the leading role.
[Related: The Latest Political Polls Collected By FiveThirtyEight]
History shows that you can largely trust these Bellwether reviews too – although you should expect the choice to bring us a few more curve balls. One or two of these states are unlikely to be so close to the national referendum; Two or three states that are not on this list are likely to turn out to be good national headlines.
The toss-up evaluations, on the other hand, should not be taken literally – at least not at this point. And that’s because handicappers can pretty well estimate how states will vote in relation to each other, but it’s more difficult to predict the national political environment so far in advance. But in a way that’s okay. The Bellwethers are really important.
It doesn’t matter if, for example, Virginia or Texas are decided with only one point, because if one of these scenarios occurs, the choice is likely to be a landslide. Knowing the Bellwethers in advance is far more valuable because we can then concentrate our energy on observing who leads them – because whoever wins the Bellwethers wins the election.