09-18-2017, 11:08 AM
Nate Silver notes some trends.
![[Image: enten-trump50states-0916-1.png?w=575&h=6...strip=info]](https://espnfivethirtyeight.files.wordpress.com/2017/09/enten-trump50states-0916-1.png?w=575&h=683&quality=90&strip=info)
Graphs belong to Nate Silver: comments are mine.
Although approval is not a 100%-reliable predictor of elections, let alone 100% precise, it is clear that President Trump is not maintaining anywhere near the support close to the vote that he got in 2016. To be sure, incumbents can usually expect to lose about 7% from the vote share entering office to approval after a significant time in office, which makes sense when you recognize that some voters can find that what they hoped for isn't on the agenda. All politicians make promises that they can never achieve; even the most honest can never predict the success of the proposals that they offer. There are, after all, nearly half the electorate and half the elected officials wanting the opposite, and they can stall just about any new legislation.
But elected pols usually get re-elected -- because they campaign again and excite much the same people the next time. The usual incumbent can show in the next election why he* was elected the first time, according to a study that Nate Silver made some years ago that I consider relevant not only to elected Governors and Senators seeking re-election but also to the Presidency. It worked well for Bill Clinton in 1996, Dubya in 2004, and Obama in 2012... and it looks likely to be relevant in 2020. Reality will dig a hole for just about any elected official, but if the hole isn't too big, he* will be able to get out of that hole.
So President Trump got 46% of the popular vote in 2016 and tracking polls typically have his approval ratings in the high 30s. Figuring that 46% of the popular vote is generally not enough with which to win a Presidential election (that is what McCain got in 2008 and Dukakis got in 1988, and less than Romney got in 2016 or Kerry in 2004, and we generally recognize them as electoral losers)... President Trump will have a tough time winning re-election.
When the mean loss is 13%, then the incumbent has a big problem unless he won 56% or more of the popular vote. The biggest losses are generally in states that he won big (losses of 17% or more from vote to approval in states that he lost only in Colorado and Minnesota), but states that were close for him (North Carolina, Arizona, Florida, Wisconsin, Pennsylvania, and Michigan which he won by margins less than 5%) are all turning on him. States on the fringe of competition for him in 2016 that he won by 5% to 9% (Georgia, Ohio, Iowa, and Texas) all give poor approval numbers.
For now I rely more upon disapproval figures, establishing a ceiling of 100 less disapproval. Disapproval is far stickier than approval, and it is a clear barrier. Effective, spirited campaigning can win over the undecided, and it can bring people to vote who otherwise would not vote. Of course some things will matter: whether the economy is or is not in the tank, whether international issues are in worse or better shape, whether there will be a discrediting scandal, whether there will be civic peace (as opposed to riots and mass demonstrations), and of course who the Democratic nominee will be and how well he* campaigns. Above all, we have no idea of whether the next Presidential election will be free and fair. The President acts much like a dictator, and there are plenty of people who would love to kill democracy so that they can have an economic order in which no human suffering is excessive so long as it enhances, indulges, and enforces class privilege. They would love to have legislation by lobbyist... forever... so long as they own the lobbyists.
Graphs belong to Nate Silver: comments are mine.
Although approval is not a 100%-reliable predictor of elections, let alone 100% precise, it is clear that President Trump is not maintaining anywhere near the support close to the vote that he got in 2016. To be sure, incumbents can usually expect to lose about 7% from the vote share entering office to approval after a significant time in office, which makes sense when you recognize that some voters can find that what they hoped for isn't on the agenda. All politicians make promises that they can never achieve; even the most honest can never predict the success of the proposals that they offer. There are, after all, nearly half the electorate and half the elected officials wanting the opposite, and they can stall just about any new legislation.
But elected pols usually get re-elected -- because they campaign again and excite much the same people the next time. The usual incumbent can show in the next election why he* was elected the first time, according to a study that Nate Silver made some years ago that I consider relevant not only to elected Governors and Senators seeking re-election but also to the Presidency. It worked well for Bill Clinton in 1996, Dubya in 2004, and Obama in 2012... and it looks likely to be relevant in 2020. Reality will dig a hole for just about any elected official, but if the hole isn't too big, he* will be able to get out of that hole.
So President Trump got 46% of the popular vote in 2016 and tracking polls typically have his approval ratings in the high 30s. Figuring that 46% of the popular vote is generally not enough with which to win a Presidential election (that is what McCain got in 2008 and Dukakis got in 1988, and less than Romney got in 2016 or Kerry in 2004, and we generally recognize them as electoral losers)... President Trump will have a tough time winning re-election.
When the mean loss is 13%, then the incumbent has a big problem unless he won 56% or more of the popular vote. The biggest losses are generally in states that he won big (losses of 17% or more from vote to approval in states that he lost only in Colorado and Minnesota), but states that were close for him (North Carolina, Arizona, Florida, Wisconsin, Pennsylvania, and Michigan which he won by margins less than 5%) are all turning on him. States on the fringe of competition for him in 2016 that he won by 5% to 9% (Georgia, Ohio, Iowa, and Texas) all give poor approval numbers.
For now I rely more upon disapproval figures, establishing a ceiling of 100 less disapproval. Disapproval is far stickier than approval, and it is a clear barrier. Effective, spirited campaigning can win over the undecided, and it can bring people to vote who otherwise would not vote. Of course some things will matter: whether the economy is or is not in the tank, whether international issues are in worse or better shape, whether there will be a discrediting scandal, whether there will be civic peace (as opposed to riots and mass demonstrations), and of course who the Democratic nominee will be and how well he* campaigns. Above all, we have no idea of whether the next Presidential election will be free and fair. The President acts much like a dictator, and there are plenty of people who would love to kill democracy so that they can have an economic order in which no human suffering is excessive so long as it enhances, indulges, and enforces class privilege. They would love to have legislation by lobbyist... forever... so long as they own the lobbyists.
The ideal subject of totalitarian rule is not the convinced Nazi or the dedicated Communist but instead the people for whom the distinction between fact and fiction, true and false, no longer exists -- Hannah Arendt.