Donald Trump has rebounded in three key swing states, according to the latest polling conducted by The New York Times and Siena. The former president now leads by five points in Arizona, four in Georgia, and two in North Carolina. 

ALSO SEE: Gallup Poll Reveals Who Is More Popular

As I'll get into, that's a major uptick from the pollster's last surveys, which were taken in late August.

AZ:
Trump: 50 (+5)
Harris: 45

GA:
Trump: 49 (+4)
Harris: 45

NC:
Trump: 49 (+3)
Harris: 47

NYT/Sienna | 9/17-9/21 pic.twitter.com/rqASf149CG

— Election Wizard (@ElectionWiz) September 23, 2024

There are two ways to analyze polling; doing anything past them is begging for issues. For one, you can look at polling averages which tend to smooth out big spikes and isolate outliers. You can also compare a poll to that same poll's prior results to spot trends. If you start "un-skewing" one poll while comparing it to a completely different poll, you might as well be throwing darts at the wall. That's why you don't see me complaining about how a poll is "weighted" and such. There's so much more that goes into a final sample than matching up partisan splits from past elections.

So how do these latest shifts compare to Times/Siena's last offerings? They had Harris up five points in Arizona and up two points in North Carolina in August. Georgia remains unchanged. That's a lot of movement in a month and could show Harris is beginning to lose momentum as we head down the final stretch. 

Naturally, the cope flowed like milk and honey. Here's current DNC hack and former Clinton lackey Simon Rosenberg.

On the NYT polls:
- PA +4, AZ -5 for Harris. Both states voted the same in 2020. 9 pts apart this time?
- Recent NYT polling has been 3-4 pts more R than overwhelming majority of polls.
- Here's most recent @MorningConsult battleground tracker: 1/https://t.co/oofKxi9WYI

— Simon Rosenberg (@SimonWDC) September 23, 2024

As I showed above, the claim that "NYT polling" has been "3-4 pts more R" is not true. Their polling favored Harris during its last round. MSNBC's "Morning Joe" also tried to dismiss the results on Monday, claiming that Times/Siena showed Mitt Romney was going to blow out Barack Obama. 

The problem? There was no Times/Siena poll in 2012.

Going back to my first point, you do not analyze polls by dismissing those that go against your narratives. The moment you start doing that, you are just guessing at who may or may not have a polling error in the current cycle. Don't like the result of a survey? Throw it in the average, but saying "but Morning Consult" isn't an argument. Times/Siena is also Nate Silver's top-rated pollster.

With all that said, Pennsylvania still looms large, with Times/Siena's last effort there showing Harris up four points. As I've been saying from the beginning of the general election, it's all going to come down to the Keystone State. Yes, you can come up with other paths to victory for both candidates, but that's mostly just fantasy. Whoever wins Pennsylvania will almost certainly win the whole thing.