It’s a safe prediction, but I’ll predict that PA is going to be the tipping point state.
I just meant a natioal emergency with no easy solution, bad economy, low popularity
far too many people would be needed in multiple states to steal/suppress that many votes. You would need an army of people wlling to commit a felony or at the least lose their jobs. So far there have been 2 instances of mail being thrown away, not sure the reason it happened, but it was only a few hundred votes and the mail was recovered. And those famouds 7 of 9 votes that were for Trump, they have been counted
Trump could win, but I am done checking the polls day after day and coming up with reasons why they may be wrong. I happened to hear Chuck Todd on the Tony Kornheiser podcast, picking football games, he said if anything, he thought the MSNBC polls are underestimating Biden. Other polls may be doing the same
Okay, okay, maybe I was exaggerating. We’ll use those poll adjustment numbers someone else pointed out above-- think they were 3.4 and 7.2-- so I’ll make it 16.6 to 12.8%.
No yard sign analysis, just an analysis of my 40 or so relatives there. 
I’m surprised ya’ll are nitpicking this prediction but not nitpicking that NC, WI and FL WILL REMAIN red.
Please realize I hope I’m very wrong about all those states. All I CAN predict is CA is going for Biden. That, I’m sure of.
Is there any reason at all to believe that they are an accurate cross-section of PA voters?
AIUI, polls that use samples of fewer than 1000 are generally considered unreliable.
The latest WI poll has Biden at +17%, and it’s a reliable pollster. One poll is not a guarantee of anything, but it’s consistent with the direction of movement in WI.
I suspect you’re right about NC and FL, though both will be close. Polls aside, both states are rife with electoral skulduggery wherever Republicans are in charge.
what specifically in FL? I still keep hoping that Biden will have a big lead 8 pm Tue, since FL is already counting mail-in votes. Not expecting it, but hoping
I believe polls of 10 million to be inaccurate. But all the relatives, their neighbors and co-workers are 100% Trump. If they’re not for Trump, they’re probably not voting, which might help.
About mail-in votes, for those that are counted (they’re counting here in CA), will we know those right away on election night? There’s, what, about 65 million cast already?
Mostly around rejection of mail-in ballots.
On a personal note, they also fucked up our mail-in ballots - the spouse and I are Florida residents who live abroad most of the year and thus vote by mail, but they sent us the “international” ballot (which only had the presidential race on it) rather than the local absentee ballot (which has every race down to Mosquito Control District Commissioner). We eventually decided to send in the ones we had, whereupon two weeks ago they sent us an “oops!” email acknowledging the problem. Thanks for nothing, Florida.
Of course it would be mail-in. Well, 700k Democrat lead at the moment, that is a lot of votes to disqualify
Have you considered that simply “polling” your relatives and their neighbors might not be a statistically valid sample size, and may actually be biased?
Do you know anything about methodology for collecting data?
Tell you what. You’ve moved your “win percentage” in PA for Trump to 16.6 to 12.8%.
I will spot you 5.8 points from your new low end estimate.
If Trump wins PA by more than 7%, I will pledge to call you an absolute genius and all around terrific person. I will bow down before you.
What difference would that make? I just hate Pennsylvanians.
And no, I don’t know how to collect poll data, I’ve never, ever been polled myself. Not even exit polls. Maybe it’s a height thing. “Nah, don’t ask him. Too short to vote.”
I think Trump is likely to win PA. Other than Philly and Pittsburgh, Pennsylvania is red Trump Country. The riots in Philadelphia have bad optics for the Democrats, especially since the guy who was shot by the police was armed with a knife and didn’t have the same helpless position of George Floyd and other unarmed men who were blatantly killed without justification. And despite Biden’s constantly name-dropping Scranton to try to emphasize his family’s blue-collar roots, most of the guys who make a living on blue collar labor are far more sympatico with Trump’s outlook on the world.
This poll is a month old but I was surprised to find that
the Harvard CAPS-Harris Poll finds that 58 percent of voters say Biden would do a better job of curbing violence in cities and 57 percent say he’d do better addressing civil unrest. Fifty-nine percent say Biden is better equipped to solve the nation’s issues on race and policing. By a margin of 54-46, Biden leads Trump on establishing law and order.
Okay, I’m hitching my wagon to this…
we can rend our hearts and garments later if I am wrong
Hasn’t it been proven that conservatives are simply less likely to respond to polls than liberals or moderates, rendering most political polling data essentially useless?
“Proven” that “most political polling data (is) essentially useless”?
No.
is the first part possibly true? If so, do pollers adjust for it?