Trump is fucked in November

RCP Spread has Biden ahead about 5.5 - don't know where your double digits number is coming from. MOE is probably +- 3.5 on these kinds of polls, assuming they reach an accurate sample (which they don't, but I'll get to that).


But as we saw in 2016, the national vote isn't what matters. It's the battleground states, which are the reason Trump won in 2016 in the first place (shoutout to Hillary for not even trying in Michigan and Wisconsin, whoopsie).

In Wisconsin, the current spread is Biden +2.7. Trump won by about 23,000 votes and .77 percent in 2016.

In Michigan, where Trump won by about 11,000 votes and .23%, the spread is Biden +5.5.

In Florida, where Trump won by 113,000 votes and 1.2%, the spread is Biden +3.3.

In North Carolina, where Trump won by 173,000 votes and 3.66%, the spread is Trump +1.

In Pennsylvania, where Trump won by 44,000 votes and .72%, the spread is Biden +6.5.

In Arizona, where Trump won by 89,000 votes and 3.5% (!!!), the spread is Biden +4.0 (easily the worst indicator for Trump. A state he solidly carried, and Biden is leading past the MOE).

In Minnesota, where Clinton won by 1.5% and 45,000 votes, there isn't a spread, but Biden looks to be ahead by about 5 points or so.


This data is not strong for Trump, but you also have to remember that the polling didn't look great for Trump in 2016, either. So let's look at it.

538, the most accurate polling company, had Trump winning about 29% of simulated elections in 2016. The reason was that the margins in all of the swing states that he won (outside of Wisconsin) gave him a reasonable shot. If a few of these fell his way, he would win - turns out all of them fell his way.

And, even Nate Silver admitted that he had allowed some subjectivity into his model because he didn't think Trump could win. So Trump's margins were a lot better than some national polls said they looked.


It's been covered pretty thoroughly that Trump voters are a little bit less likely to say their preferences for polling, which results in a slight underestimate of Trump's position. It's not as big as some would have you believe, but it can be as large as a percentage point.

If we give Trump +1 in every state, only Michigan and Pennslyvania fall outside of the typical MOE. That's 36 electoral votes to Biden. Reminder that there were 7 faithless electors in 2016 - Trump's total by win was 306, and Clinton's was 232. That gives Trump 270-Biden 268, assuming nothing changes.

But it isn't that simple.

Campaigning hasn't even started yet. Biden is very, very vulnerable to lots of attacks - the Democratic party base simply is not unified under Biden the way the Republican party is under Trump. If the exact same amount of attack ads ran on Biden as they did on Trump, Biden would lose more voters.

Based on how the Democratic primary played out in terms of Biden losing steam, it's pretty safe to think the margin will be under 3 points by election day.

In other words, although Biden will probably win the popular vote based on what we see right now, the electoral college will once again likely play in Trump's favor.

My eyes have rolled at poles of all shapes and sizes since 2016. Maybe they can find an algorithm to control for what happens when the booth curtain closes and people can vote their minds rather than what they can get away with saying aloud in their social circles.

And we’re a long way from November. Still VPs to pick. Lots of campaigning to do, and as much time as there is for Trump to show his ass, it’s very possible we see Biden have a medical event on the debate stage or elsewhere.

Great TV ahead.