Biden beat Trump by 4.4% in the popular vote; a fairly decent margin. However, the Electoral College race actually came disastrously close to defeat; Biden only beat Trump by a margin of about 43,000 votes in Arizona, Georgia and Wisconsin - combined in total. That’s less than half a football stadium’s worth of voters. Had those three states gone red instead of blue, Trump would still be president today.
What that means is that, had Biden beaten Trump by only 4.0% in the popular vote instead of 4.4%, it’s possible that that would have had some sort of a-descending-tide-sinks-all-boats effect that might have caused that incredibly slim 43,000-vote margin to vanish - and with it, Biden’s presidency.
To make matters worse yet for the Democrats, the last election didn’t involve Census re-allotment. This time around (2024,) the Electoral College has been readjusted to reflect population shifts, and it resulted in a Republican net gain. Someone correct me if I’m wrong, but I believe Trump states gained 4 EVs and Biden states lost 4 EVs. So this means the Electoral College favors Republicans even more today than four years ago.
So as a general rule of thumb, can it be said that the Democrat must beat the Republican by at least 4% in the popular vote to stand an equal chance of winning the presidency, and in fact that only a margin of 5% or greater in the popular vote means a truly-favored chance of taking the EC?
The two are not really all that related. If you win a state by 50%, you get it’s EC votes. If you win by 1%, you get the EC votes. So, someone could pile up a huge popular vote count in sure thing states, but fail in purple states.
Earlier in the year I was hearing 3-4% overperformance in the popular vote was going to be a minimum necessary to defeat Trump based on election results from 2016 and 2020. However, recently I’d been hearing that the gap was shrinking, in part because of Latino & Black defections to Trump. So the Democrats would still win states like CA, NY & IL but would win them by slimmer (though still comfortable) margins thus meaning the popular vote margin would be slimmer. And they would lose states like Texas and Florida by wider margins which wouldn’t change the Electoral College results but would run up Trump’s popular vote totals. This didn’t seem to apply as much to the big swing states (perhaps minority populations there were less likely to risk a Trump presidency in the polling).
But now, with Harris in the running and some of the defections returning to the Democratic camp, we might be back closer to margins like in 2016/2020. But I don’t think anyone really knows yet.
Theoretically, yes. But practically, no. No real candidate would ever get 51% in one set of states and 0% in the other. Different states are going to be correlated, so it’s meaningful to put a number on what the OP is asking.
Sure, but again, it’s overall correlated. It’s not likely that Kamala would get a massive 20% jump in votes in California while picking up only an extra 1% in Florida and Michigan.
Hillary lost in 2016 despite winning the popular vote by 2%.
Biden won the election in 2020 with a 4.5% popular vote win, despite it being perilously close in a few key battleground states.
I would say Harris might need anywhere from a 5-6% popular vote win to win the electoral college as well, at least in a decisive manner. It’s a bit harder than it was in 2020 since the census that year gave red states a few more EVs (and took them from blue states).
(In case it’s not clear, I’m pretty much agreeing with your assessment in the OP.)
Sort of. Not in a simple linear fashion though. Certainly not all states correlated together equally though. Like states, demographically tend to travel together. And of course specific states respond to specific issues, and to the investments campaigns make in their states (money, visits, GOTV infrastructure, etc.)
I can’t find any data on this myself: there has been significant movement to more rural locations associated with remote work beginning during Covid. Has that impacted voter demographics in any meaningful way?
The presidential numbers don’t really transfer well to the electoral college, but they do have some gravity on the lower races. A big swing to Harris means more Democrats voting. That would make no difference to her winning New York State, e.g. But there are several Congressional seats that went to Republicans that could swing back if the turnout was big for Democrats.
RCP rates seven Senate seats as toss-ups, five of them in “battleground” states. Turnout would surely make a difference in some of those. And the Dems need to win all seven to keep the Senate. Doing so was thought of as next to impossible, but might, maybe, possibly, could happen with a blue wave.
It’s not, though. If Harris wins exactly the same states, she still wins the election. The likelihood the reapportioning would make the difference for Trump is very low.
True, but again, Biden only won in 2020 because of a sliver’s margin in three states - Arizona, Georgia and Wisconsin - just 43,000 votes in combined margin. Without those three states, the election would have been a 269-269 tie in the Electoral College and thrown to the House - but that depended on every single Republican elector remaining faithful.
Given the change in reapportionment, if a tiny downturn in Democratic vote caused Kamala to lose those three states by a tiny sliver, it wouldn’t be a 269-269 tie - it would be an outright 273-265 Republican victory. Which is probably enough to prevent faithless Republican electors from gumming up the works.
That article really points us to the data driven way to answer the question: how much under national average are the likely tipping point states running? IOW what’s the delta?
Right now looking at the NYT polling averages Harris is nationally up one point, and is up a point to two points in the Blue Wall states, down two in Georgia.
So at point in time the likeliest tipping point states that are essentially tracking with the national average, which would mean that a national average win should translate to an EC win.
Will there be more divergence as more polling comes in? Maybe but I wouldn’t bet on which way if so.
Of course I am open to the possibility of systemic errors in polling also.
In the opinion of one random Internet nerd Harris will need a popular vote advantage of about 3.3% to be a slight favorite to win the EC.
Since we won’t know the popular vote result until after we know the Electoral counts, perhaps more relevant is the relationship between nationwide polls in late October and the popular vote. Biden was +8 in the final polls but only +4.5 in the election. Hillary was about +4 in the final 2016 polls, but she won the popular vote by only +2.1. Apparently many Blue voters stayed home on Election Day. I hope this year it’s the Reds who stay home.
We can take random internet dude’s statement FWIW.
As to how close final polls come to election results: Obama Election Days were +2.9 over Romney and -1.6 under polling with McCain. Since he won both no one really cared much. Off course these are poll results without the footwork that the analysts like Silver apply.
Still being off by 2 to even 4 in one direction or the other, and it has gone both ways, seems to be a reasonable expectation.
And while EC advantage seems fairly small right now, per polling anyway, it’s probably reasonable to bank on a two point one anyway. So give me a six point advantage and I will be a bit less anxious.
I don’t think there’s much reason to think that. The history suggests Democratic voters are less motivated, at least where Trump is concerned.
The deep, deep love for Trump is underestimated in these circles. He is much more hero-worshipped than any other living American politician. As it stands, Harris is PROBABLY losing. Not certainly, but Trump would be likelier to win if the election were held this week. Maybe two chances in three.
I’ve heard a few times that the figure the OP is looking for is 3%. This is an approximation, though – it’s not as though multiple elections at 2.9% are sure to go one way and multiple elections at 3.1% are sure to go the other. I think the utility of it is in quickly making sense of yet another national pole. Whereas in democracies without an EC, a polling majority is the tipping point for guessing what it means, in our country the tipping point is D+3.