Tag Archives: polls

National

Penultimate national election analysis

Election Day starts in less than 77 hours. Barring some significant event over the weekend, this will be our penultimate election analysis. A series of fresh polls have come out at the national and state level, showing little change in the race. President Trump is not speaking, tweeting, or positioning himself as a man confident he can win on Tuesday.

Firewall States: We consider firewall states as the state on the FiveThirtyEight Snake Chart as having a 90% chance, or better, of tipping to a candidate. All of the states above that state (Trump) or below that state (Biden) are not battleground states.

Trump’s firewall state of South Carolina remains unchanged. Biden’s firewall state has changed from Wisconsin to Nevada. This change happened because Biden’s position has strengthened in both states.

Battleground States: We consider the following as battleground states in order of their lean from Trump to Biden:

State or District

538

RCP

Malcontent
News

Texas

Trump +2

Trump +2.3

Trump 

Iowa 

Trump +0.7

Biden +1.2

Toss Up

Ohio 

Trump +0.6

Tie

Toss Up 

Georgia 

Biden +1.0

Biden +0.8

Toss Up

Maine 2nd District

Biden +0.2

Not Tracked

Biden 

North Carolina

Biden +2.1

Biden +1.2

Toss Up 

Florida

Biden +2.1

Biden +1.2

Toss Up

Arizona

Biden +2.8

Tie

Toss Up 

Nebraska 2nd District

Biden +4.5

Not Tracked 

Biden 

Pennsylvania

Biden +5.1 

Biden +3.6

Biden 

As we head into the final weekend before the election, Iowa, Ohio, Georgia, North Carolina, Florida, and Arizona are the critical states for both candidates. It is worth noting that Trump carried all of these states in 2016. Although Real Clear Politics considers Wisconsin and Minnesota as battleground states, data indicates neither state is in play. We have consistently said that where New Hampshire goes, Maine District 2 goes and data indicates that the district is strengthening for Biden. 

There are some other election and polling trends worth noting. More early voting ballots have been submitted in Texas than all the votes received in the 2016 election. Iowa has announced they are closing 267 polling sites, mostly in the metro-Des Moines area, due to COVID. There are concerns this change will disenfranchise Democratic voters in the blue-leaning city. Additionally, the Supreme Court has decided that North Carolina and Pennsylvania will need to accept ballots postmarked by election day. However, the courts have also decided that Pennsylvania can’t start counting the over 2-million early ballots it received until election day. As a precursor to election chaos, over 230 election-related cases have been filed in federal courts across the country.

On Tuesday, Trump suggested that people could change their votes they’ve already submitted, creating the topic to trend on Google. A 45 state analysis indicated that only a handful of states allow “spoiled ballots” and only under certain conditions. Further, the window to change a ballot has closed in every state except New Hampshire.  

But what if the polls are wrong like in 2016. We’ve covered previously on who 2020 is different; most critically, Biden has consistently polled higher than Hilary Clinton did in 2016. There has not been the same indication of deflating poll numbers that battered Clinton in 2016, and we’re past the window of a Comney memo grade surprise. If the same polling error in 2016 of about 3% were to happen in 2020, Biden would still win based on all available data. 

We believe the more significant question is what President Trump will do post-election. Suppose Trump moves to invalidate the election or manipulate electors. In that case, it will almost certainly be met with protests. Federal, state, and local police officials and business leaders are very worried about post-election violence. In numerous communities, both small and large, businesses have already started boarding up windows and taking other defensive measures.  

Neither candidate nor their supporters should be thinking the election is already over. Fivethirtyeight forecasts an 89% chance Biden will win the national election. By our analysis, the race has improved for Biden to 308 vs. 230, Biden win. Real Clear Politics no toss-up map also shows the race improving for Biden. As of this writing, they are indicating the final results of 345 vs. 193, Biden.

National

The election dashboard flashes yellow for Biden

A slew of new polls has come out over the weekend. Although they don’t indicate the Democrats should panic, they should be moving to yellow alert.

In deciphering the data coming out, each election cycle poses challenges when reading between the lines. In 2016, pollsters missed the divide between white non-college-educated and white college-educated males, with the latter making up a significant portion of Trump’s base. Although many are quick to dismiss the pollsters are egregiously wrong in 2016, the reality is most pollsters came within the margin of error. In the four key states that gave Trump the win, the margin of victory was narrow.

The so-called firewall state for Biden shifted back to Nevada over the weekend, while it has remained South Carolina for Trump. We consider a firewall state a state where there is a 90% or better chance that the state will go to a particular candidate. Any state “left” or “right” of that state (depending on the political leanings) is a lock unless something incredible and unforeseen happens.

Although several bad polls out of Texas battered Trump, and early voting numbers there are massive and giving Biden a considerable advantage, the latest Sienna/NYT poll has Trump up by 4 this morning. Trump grew his tenuous leads in Iowa, Georgia, and Ohio over the weekend.

For Biden, the leads he is holding in North Carolina, Florida, Arizona, and Pennsylvania contracted. If there is a repeat of 2016 polling errors, all three of these states would be won by Trump. A couple of polls out of Arizona show a wild swing in the Kelly(D) vs. McSally (R-I) race. I would consider these numbers suspect as a 13-point swing is hard to process with no apparent event in the race to support it.

Another point of concern with Democratic strategists is the Biden team is turning their focus to Texas and Georgia, believing they can flip these states. Some see this as a repeat of 2016 and want the Biden campaign to prioritize defending the lead in the rust belt states and Pennsylvania while working on flipping Iowa and Ohio.

For Biden supporters, they shouldn’t be hitting the panic button – yet. Biden holds an aggregate national lead of 8-1/2% to 10%, which is significantly larger than Clinton at this same point in 2016 (5.6%) and well outside the margin of error. Additionally, one factor that pollsters may not be correcting enough for is the dwindling number of likely voters who are willing to respond to a poll and haven’t voted yet. Some models predict 100 million Americas will have voted before election day, which is close to 80% of all the votes case in 2016. Are the polls over counting the people who haven’t voted yet, leaning more toward Trump? Pollsters are undoubtedly aware of this and adjusting their models, but do they have a reliable baseline to use?

Regardless of your political affiliation, the most critical thing you can do is vote and vote early. With officials concerned about election day violence, voter intimidation, and ballot boxes burned in numerous cities, there is sure to be manufactured chaos on election day. If you want your vote to count, then do it early this week if possible.