ADVERTISEMENT

So who is willing to accept

538 might want to downgrade NYT from their a+ rating.

I think a lot of that rating has to do with methodology and transparency as well.

Look at Trafalgar - do they have some secret sauce in the Trump era? Maybe - but they black box it so there's no way to know. Could also be that their partisan bias is the perfect antidote to a systemic inability of pollsters to accurately stratify the electorate in the current environment.

I mean let's face it - if you polled like a regular pollster and then just fudged everything by 4 pts or so in Trump's favor you'd look like genius.

I'd *really* love to see what the internal campaign data modeling looks like with the financial resources they have. If we're to believe the slides floating around from Biden's campaign yesterday, they had the race super tight in these states. Per Twitter, Trump's internal polling had the 3 rust belt states within a point also.

Which would suggest the campaigns knew the outlook much better than public pollsters did. Why is that and why can't we replicate that publicly?
 
  • Like
Reactions: ChrisKnight06
I think a lot of that rating has to do with methodology and transparency as well.

Look at Trafalgar - do they have some secret sauce in the Trump era? Maybe - but they black box it so there's no way to know. Could also be that their partisan bias is the perfect antidote to a systemic inability of pollsters to accurately stratify the electorate in the current environment.

I mean let's face it - if you polled like a regular pollster and then just fudged everything by 4 pts or so in Trump's favor you'd look like genius.

I'd *really* love to see what the internal campaign data modeling looks like with the financial resources they have. If we're to believe the slides floating around from Biden's campaign yesterday, they had the race super tight in these states. Per Twitter, Trump's internal polling had the 3 rust belt states within a point also.

Which would suggest the campaigns knew the outlook much better than public pollsters did. Why is that and why can't we replicate that publicly?
Good question. Supposedly internal polling is always more accurate. If that's true, it makes sense why Trump is pissed.
 
Good question. Supposedly internal polling is always more accurate. If that's true, it makes sense why Trump is pissed.

Interesting also cause I've seen lots of stuff that folks on both sides are pissed because the polling being so far off meant they were putting money in the wrong races.

Wasserman on Twitter said that district level polling has never been this off before. So no idea how the national campaigns got it right if everyone else was wrong.

Theory - the national campaigns were off too, but they biased their numbers as needed to simulate a toss up and made resource allocation decisions based on that. Either that or some really smart dudes built out some really good models using a bunch of non-polling data.
 
Interesting also cause I've seen lots of stuff that folks on both sides are pissed because the polling being so far off meant they were putting money in the wrong races.

Wasserman on Twitter said that district level polling has never been this off before. So no idea how the national campaigns got it right if everyone else was wrong.

Theory - the national campaigns were off too, but they biased their numbers as needed to simulate a toss up and made resource allocation decisions based on that. Either that or some really smart dudes built out some really good models using a bunch of non-polling data.
I've heard that too, specifically about the Lindsey Graham and Mitch McConnell races.

Polling aside, I dont understand how the rust belt actually turned out the way it did. You would think an 8 point Trump win in Ohio would be kind of a bellwether and it would mean at least getting 1 more win, but somehow its an outlier. Its also strange how Trump improved in New York by 12 points but lost ground in California. How did Trump improve so much in border towns in Texas that have high Hispanic populations but lost ground in New Mexico? The final analysis of why things went the way they have is going to be interesting.
 
I've heard that too, specifically about the Lindsey Graham and Mitch McConnell races.

Polling aside, I dont understand how the rust belt actually turned out the way it did. You would think an 8 point Trump win in Ohio would be kind of a bellwether and it would mean at least getting 1 more win, but somehow its an outlier. Its also strange how Trump improved in New York by 12 points but lost ground in California. How did Trump improve so much in border towns in Texas that have high Hispanic populations but lost ground in New Mexico? The final analysis of why things went the way they have is going to be interesting.

Ohio is not a bellweather for the rest of the Rust belt. Ohio went red in 04 and 00 while MI, WI, and PA all went blue. The rust belt is essentially blue states, with Ohio being a swing state, and Indiana (if it counts as rust belt) being red. 2016 was obviously an exception when all of those states went red, but historically speaking, Ohio going red does not mean the other states are going red too.
 
Ohio is not a bellweather for the rest of the Rust belt. Ohio went red in 04 and 00 while MI, WI, and PA all went blue. The rust belt is essentially blue states, with Ohio being a swing state, and Indiana (if it counts as rust belt) being red. 2016 was obviously an exception when all of those states went red, but historically speaking, Ohio going red does not mean the other states are going red too.
I was referring more to the margin of victory being that high.
 
I was referring more to the margin of victory being that high.

But it still doesnt mean much of anything. We will see going forward if 2016 was an anomaly, or if those states are true swing states at this point. But from a recent historical perspective, Ohio going red isnt indicative at all with how the other states vote.
 
ADVERTISEMENT
ADVERTISEMENT