Originally Posted by Rod Kaufman
"Silver is not and never has been a pollster."
He's evidently described as such by the media, "Per the pollster's forecast, Harris has a 56.7 percent chance of winning the necessary 270 electoral votes in November, while Trump trails with 43.1 percent."
https://www.newsweek.com/nate-silvers-swing-state-forecast-shows-new-movement-1963629
BTW: I still think Trump will prevail, you?
Hi Rod:
Yesterday, I would have agreed. But today the Biden/Harris administration has apparently resolved the longshoreman strike. This is a big deal for Harris. A major supply chain disruption peaking around election day would have been bad. So today I think Harris wins.
Here is something I find fascinating. Silver ran his model 70,000 times earlier this week to quantify the most probable outcomes. As you know the overall is Harris winning the EC 56% of the time and Trump 44% of the time which is nearly a toss-up. Silver's model applies possible polling errors, such as one candidate being under polled or certain demographics being under polled and so on. There are always some polling errors, often 2-3%. Also, the swing states can be correlated error wise, this is what happened in 2016 when Trump was under-polled across the board and everyone predicted that Clinton would win. The polls were not off by that much, but they were all off in the same direction.
Anyway, the most likely scenarios in Silver's recent 70,000 run were: (A) Harris sweeps all 7 of the swing states, and (B) Trump sweeps all 7 of the swing states. Together those two outcomes accounted for 40% of all results. This is another way of saying that the polling is entirely within the margin of error, plus observing that the errors are often correlated.
Everyone knows Trump was under-polled in '16 and '20. If this is happening again, Trump wins for sure. On the other hand, the better pollsters are certainly trying to fix whatever caused Trump to under-poll in '16 and '20. If they over-correct, Harris wins easily. If they are right this time though, and several swing states are within 1% on election day, look for chaos and allegations (i.e. Russia, voting machines) on both sides and no final determination for weeks.