“Florida shattered its opening day record for in-person early voting Monday, with at least 350,000 people casting ballots and election officials continuing to count statewide late into the night.”
- Politico reports.
We are now exactly two weeks out from November 3rd and the polling is not turning for the GOP. In fact, today we find out that in the one area where voters were giving Trump higher marks than not – the economy – voters now disapprove of him. A new Financial Times poll finds that 46% of voters now think Trump has done more to hurt the economy with 44% holding the opposite view. This is another data point that shows the electorate continuing to trend away from Trump and the Republicans. That said, Biden’s lead in the FiveThirtyEight polling average is down today to plus 10.4 points. This does not suggest that Biden is losing support. To the contrary, Biden’s lead over the past week or so is fairly consistent around plus 10.5 points.
How is the Senate shaping up?
Historically, the Toss-Up Senate seats go to one party by 70-80%. The one exception was in 2016 when where the GOP got just 56% of the Toss-Ups. Cook Political Report lists seven Senate seats as Toss-Ups. All are currently held by Republicans.
According to yesterday’s CPR briefing, it will be announced today that Iowa is moving from Toss-Up to Lean Democrat. Larry Sabato’s Crystal Ball has already made the move. Democrat Theresa Greenfield continues to build momentum. Last week’s debate probably did not help Senator Joni Ernst (R) as the incumbent who won her first term describing herself as a farmer had no idea how much soybeans cost while Greenfield was able to correctly explain the price of corn. If you remember back a few weeks to our discussion of whether debates matter you’ll know that one way they make a difference is when one candidate makes a big gaffe that either underscores or dismantles a narrative about a candidate. Soybeans and corn might not make for interesting debate topics in many states, but they are important economic issues for Iowa. That Farmer Joni had no idea how much soybeans cost may have helped to dismantle her own narrative. Couple that with Greenfield’s success it might also suggest to voters that Ernst is the elitist in the race (which is a play Republicans consistently try to make on Democratic opponents).
Recently, CPR moved Loeffler’s (R-GA) seat to Toss-Up, Sullivan’s (R-AK) to Lean Republican from Likely Republican, and the same for Cornyn (R-TX). Sullivan has since seen one poll showing his opponent Al Gross (I-AK) beating him, but another from Siena College (A+) showing the Republican in a comfortable lead. Alaska has a reputation for being difficult to survey, so a Lean Republican projection is probably fair enough. Expect almost anything to happen in this race from a Sullivan blow-out to a clear, but close Gross victory.
It’s not clear what is going on in Texas right now, except that Cornyn is concerned. Recent polling has shown Democrat MJ Hegar closing the gap, but still losing by single digits. Yet just this week Cornyn announced that he has disagreed with Trump on some issues – including building the border wall. He says he did it privately, which is why no one ever heard him being critical of Trump until two weeks before he’s up for reelection. In the past I’ve noted that you can learn a lot about how campaigns are doing based on what they do and how they spend resources. Cornyn’s recent comments suggest that he is concerned about what their internal polling is telling them. He must be losing independents and Democrats that previously supported him; otherwise, it is not clear why he would raise the issue of opposing Trump on the border wall at this late stage. The borderlands in Texas are Democratic territory and opposing the wall is not something that appeals to Cornyn’s base. Texas might be the most interesting state to watch in two weeks.
For a long time in American history, vote splitting was common (although it was less common in the 19th century). Recent partisanship brought on by Gingrich’s 1994 scorched earth campaign to win the House and install him as Speaker either changed that or simply coincided with a new phenomenon of single party voting. When it comes to Senate elections, Pew has found that since 1988 most of these races go the way the state votes for president.
As you can see from the chart above, since 1988 the share of Senate elections won by the same party as the winner of the state’s presidential vote has increased over time. At this point, looking at the presidential vote share (determined via polling) in each state appears to be a useful indicator for Senate projections. Let’s apply the logic to current presidential polling in states with competitive Senate elections this year.
If the presidential polling is accurate and holds over the next two weeks, Pew’s research suggests that the Democrats would pick up seven Senate seats and lose one (Alabama) for a net of six. This would change the make-up of the Senate from 53-47 GOP to 53-47 Democratic. In yesterday’s CPR briefing analysts thought that a 53 or 54 majority for Senate Democrats was becoming increasing likely (although they maintain that 51 is a safer projection).
There has been fewer polling over the past few days than previously. Most likely this is again related to the fact that polling is expensive. Most polling sponsors want polls out in the last week of the campaign and if they need to cut back based on cost it will be mid-October that suffers. Expect to see a lot of polling next week - and probably not much for the rest of this week.