In recent weeks, with a spate of optimistic polling for Democrats and Trump aligned Republicans winning their primaries (the love for Liz Cheney is optimistic considering the media fed the meme of Dick Cheney being the Emperor from Star Wars) has given the party renewed optimism.

This optimism is not just fed by polling and candidate quality but also legislative successes (well, what they call successes). First, Democrats passed with a majority of GOP support a bipartisanship CHIPS bill. Then, right after, Democrats unveiled their reconciliation passed Inflation Reduction Act (Note: It just passed and no respectable economist expects it to do what it says). Then, Democrats helped shuttle through a bi-partisan Veterans Bill to rightly support soldiers adversely impacted by Burn Pits from the Iraq War.

All this has the party looking up. I mean, the polls are pointing to a blue wave if you believe them in swing states and of course Dobbs will make suburban voters turn out in droves for the party.

Except for the albatross in the room; Joe freaking Biden. Despite all the legislative successes the President has enjoyed he is still mired in the low 40’s per both RCP and 538. That is a major problem because history tells us two things; when an incumbent president is unpopular his party loses. And even when he is popular they CAN lose.

History is the evidence. There have been one dozen midterm elections from 1974. In only two of them has the incumbent President’s party gained seats; what some call asterisk elections (more on that in a moment). Both elections where the President’s party gained seats were extremely unique.

In 1998, the GOP was overstretched due to its gains from 1994 and few losses in 96, and they went after a President who had an approval of over 66 percent. The result was a loss of five seats (arguably they should have lost more). In 2002, George Bush benefited from a Rally Around the Flag effect and with an approval of 63 percent his party picked up a mere six seats.

But, here is the rub too, even a President with a HIGH approval can still lose seats. It happened to Ronald Reagan in 1986 when his party lost five seats and he had a 63 percent approval rating. It also happened to HW Bush in 1990 where his party lost eight seats even as he boasted a 58 percent approval rating.

In short, this means high approvals help Presidents avoid blowbacks but they don’t stop them from losing seats. Now consider despite his recent legislative successes where President Biden sits. Per RCP and 538, President Biden sits at less than 41 percent approval. Even considering some Democrats disapprove and are unlikely to vote for Republican candidates, that is pretty bad. Worse, just historically, only President Truman and Jimmy Carter had lower approval ratings at this point in their Presidency.

Can Biden up his approval rating to help his party? Unlikely.  Per Nathan Gonzales, “Looking back more than 70 years, there has not been a single president who substantially improved his job approval rating from late January/February of a midterm election year to late October/early November. In the last 18 midterm elections going back to Harry Truman in 1950, the average president’s job approval rating dropped eight points between this time of year and election day.”

But, let’s suppose Biden’s approval goes up even more from its low of around 36 percent and remakes history, then what? Well, probably not much. First – off, to repeat, it probably won’t happen. Biden has dropped in the last week in the RCP average from 41.2 percent to 40.8 percent and this is after he just had the most recent spate of legislative success in his Presidency.

Secondly, history just does not tell us a higher approval of say 45 to 50 percent will mean much. Barack Obama had an approval rating of 45 percent and lost 63 seats (partially due to Democratic gains in 2006 and 2008). Even President Clinton was in the 40’s when 1994 rolled around.

So, I hate to tell Democratic cheerleaders like Simon Rosenberg, but it is not a newer, blue election. Particular to House elections, it is notable in internal surveys for DEMOCRATS a Republican like Myra Flores is running only four points behind in a D + 14 district. Or how about Jaime Herrell in NM who is running a single point behind in a D + 6 district. Heck, even Republicans in suburban districts like Don Bacon are outrunning the partisanship of their seats.

Then, there is the large number of undecided voters on the generic ballot test. Contrast where Democrats were in 2020 at this point (where they won by 3 points when polls said they would win by an average of 7) with now. On August 19th, 2020, Democrats were averaging over 48 percent on the generic ballot test (darn close to 50 percent). Today, they are averaging 43.9 percent. That’s a stunning drop. Heck, even Republicans are running stronger now than than by 2 points. Combined with the fact almost NEVER have undecided voters broke strongly for the incumbent party with an unpopular President and you can see this optimism is founded on a house of cards.

So, what do we make of the Senate? I mean, after all, Rosenburg and CNN are telling us how GOP candidate quality is hampering them. Numerous surveys have come out showing Democrats running strong or Republicans surprisingly far behind. And, of course, there are the comments from Mitch McConnell basically saying he would be happier with a Democratic Senate than one with Republican Senators who won’t do what he wants (I am paraphrasing).

I made a couple of these points in a prior post but its worth repeating. While these Democratic Senators and candidates have stronger favorable ratings and approval then the President history again suggests that only goes so far. Since 2010, incumbent Senators or candidates of the incumbent party in the White House have been able to outrun the President’s approval by about five points. That does not bode well.

Much has been made of just how unpopular Republican candidates are in states like Arizona, Georgia, and Pennsylvania. But again, even if you use Harry Enten’s argument at CNN that voters usually support the candidate they view more favorably, you ignore the datapoint that a solid majority of voters think the country is on the wrong track and are those voters really going to support the party leading the country?

Further, we saw this as far back as 2014, Republicans are more undecided than Democrats in surveys. Are these Republicans really going to support the Democratic candidate for Senator? Probably not.

Lastly, we come to the elephant in the room no polling analyst (who makes their livelihood on this stuff) will admit, but polling is struggling. It has struggled since 2012 and routinely leans left since 2014 (in 2012 it leaned rightward at the state level).

Again, to be fair, I do not think there is some grand conspiracy of pollsters who have an agenda. But, I do think, they have a limited set of tools at their disposal to gauge public opinion. With the reshuffling of the party’s coalitions where upscale, suburban voters are more Democratic and likely to answer pollsters, while the average GOP has become more downscale less prone to answer what they deem “elites” queries and it is a recipe for pollsters to miss again and again and again. To some extent, this has happened repeatedly in states like Wisconsin, Florida, and Ohio, where the party realignment has been strong among blue – collar whites and Hispanics. As one analyst put it after 2018 in Ohio, “the parties and pollsters don’t know who they are talking to.”

This leaves pollsters with little option but to weight their samples according to their electoral expectations and that makes them biased towards weighting them along political headlines. Or you could just be Yougov which has a sample of 36D/25R/39I which is absolutely insane considering the 2020 electorate in a D+4 environment did not look like that.

It’s likely Republicans won’t sweep every competitive Senate or statewide gubernatorial race. That’s just how elections work. It’s likely their House gains will be limited by Dobbs in suburban, D leaning districts and redistricting.

To wrap this article up, as David Drucker said, “Midterm cycles since ’06 have certain rhythm,” Drucker wrote. “1) Maybe POTUS’ party will avoid losses. 2) Things look good for the out party. 3) Things look REALLY good for the out party. 4) Hold on, maybe POTUS’ party won’t lose as many seats as thought. 5) Could POTUS’ party avoid wipeout? 6) WIPEPOUT.”