In recent months, the poll numbers coming out of Arizona have been stunning. Joe Biden maintains a consistent lead over Donald Trump, and Mark Kelly continues to blow away Martha McSally. But how did this happen? How did the land of John McCain and Barry Goldwater – a state Trump won by 3.5% – become one of Democrats’ best flip opportunities? The answer lies in the results and turnout dynamics from 2018.
The Senate Race
Perhaps the most shocking part of this cycle has been how far ahead Mark Kelly has been polling of Martha McSally. While he was always considered a strong candidate, his double digit lead in most polls has caught observers off guard. But, this should not be surprising given the underlying dynamics.
First of all, it is worth examining why the 2018 results in Arizona may be more predictive than most midterms. Midterms are usually poor indicators of the next presidential election for a few reasons, foremost among them being turnout. Winning parties often benefit from a turnout advantage that cannot be sustained into a presidential election year. But did this happen in Arizona in 2018? The answer is no. A precinct level analysis of turnout data shows that there was virtually no correlation between Hillary Clinton’s vote share in a precinct and how much the total votes declined between 2016 and 2018. In fact, any correlation that exists is slightly negative, meaning Democratic areas declined more in votes cast than Republican ones. Below is a map of how this played out by congressional district.
As can be seen, the number of votes cast declined the most in heavily Democratic districts, not heavily Republican ones. This has major implications for 2020. If the electorate was largely the same between 2016 and 2018, there is no reason to believe the group of voters coming to the polls in 2020 will be much better for the GOP than the one that gave Democrats a U.S. Senate seat.
Of course, just because the electorate is the same doesn’t mean the national environment has to be. The second reason why midterms are unrepresentative is because voter attitudes change. Many politicians have been swept away in a wave election, only to be brought back in by the tide two years later. The bad news for McSally, however, is that the national environment doesn’t seem to be reverting. The generic congressional ballot, according to FiveThirtyEight, sits at D +8%, hardly changed from 2018’s final margin of D +8.5%. Additionally, that final margin does not take into account races uncontested by Republicans, of which there were many.
Given that, according to RealClear Politics, Joe Biden holds an 8% advantage over President Trump in general election polls; it does not look like as of today that the environment has gotten much redder from 2018. The assumption that the environment would be more favorable to Republicans this cycle was a major driving factor in much of the GOP’s optimism.
On top of this, McSally has her own issues as a candidate. Her appointment to the seat after a close but decisive loss to Kyrsten Sinema, no doubt left a bitter taste in some voters’ mouths. And as outlined above, she will be facing virtually the same electorate that sent her home last time. There is no cohort of Trump voters ready to flood into the voting booths and counteract the negative impact her appointment may have had. Additionally, McSally’s performance in 2018 was not as impressive as the conventional wisdom holds. Yes, it was a close loss, but she was the second worst performing Republican on the ballot. She even underperformed the GOP’s U.S. House candidates, who left an entire district uncontested.
McSally’s lackluster 2018 run, the lack of elasticity in electorate makeup, the national environment, and the situation surrounding her appointment to the seat have all combined to make it clear – she’s not starting from a strong position. The question has now turned to how she can fix this. She needs to make meaningful inroads with voters that rejected her last time.
This is no small task for any politician, but McSally does not seem to be trying very hard to accomplish it. She has tied herself more closely to the President than ever before, portraying a more conservative image. She called a CNN reporter a “liberal hack,” and fundraised off the incident. In February, she appeared at a rally with the President, where he said McSally had his “complete and total support”. McSally followed this up by appearing on stage and saying that she had a message for the “liberal hack media in the back,” and saying the President’s State of the Union “brought tears to my eyes”. But is this a prudent strategy? An examination of the evidence states probably not.
The Senate race saw more votes cast than other contests in Arizona. Gubernatorial, U.S. House, row offices, the Senate race beat them all. Voters were not skipping out on voting in the election. This makes it seem unlikely that McSally lost because Trump voters were sitting out the race. According to CNN, she won voters who approved of Trump 87%-11%. Those who disapproved broke for Sinema 94%-5%. It is possible that had McSally won Trump approvers by this margin, she would have eked out a narrow victory. But shifting her entire image in an effort to sway the 1/20th of the electorate that approved of Trump but voted for Sinema is of questionable wisdom.
In all likelihood, this pivot alienates more voters than it wins. And it is not as if the President’s support of her in 2018 was a closely guarded secret. It may be simply that Sinema was a good candidate who was capable of winning crossover votes. Mark Kelly appears to be a candidate with similar abilities. Recent polls show him having crossover appeal similar to Sinema’s, even after McSally’s appointment and pivot.
The Presidential Race
The race for the White House in the Grand Canyon State mirrors many of the dynamics found in the Senate race. The electorate will be largely the same as the one that showed up in 2018, and the national environment has not yet shifted substantially in favor of the GOP. Granted, Biden seems to have less crossover support than Kelly, and Trump does not have some of McSally’s issues.
But nonetheless, Joe Biden has pulled out to a solid lead in the state’s polling average. While the size and consistency of this lead has caught many off guard, it is not as shocking when looking at how the state has trended. Mitt Romney won the state by 9%, and Donald Trump won it by 3.5%, even as the national environment swung 2% in favor of Republicans from 2012-2016. That means Arizona trended 7.5% to the left over those four years, a rapid shift. But this was largely glossed over by pundits.
The media’s infatuation with Trump’s Midwestern success meant they paid less attention to states that shifted left such as Arizona. Furthermore, the heavily mail-ballot nature of Arizona’s elections meant the final margin was not known for days or weeks. At this point it would have been old news. A completely secondary story to Trump’s overall victory. The numbers made it clear that Arizona would be a top Democratic target in 2020, but because Arizona has been known as nothing except a Republican stronghold for years, some may find it difficult to envision Arizona actually flipping.
It all feeds back to the Senate race and why Kelly’s lead is so large. If Trump is losing the state or nearly tied, then going all in on him becomes an even riskier strategy for McSally. And the longer she attempts to maintain her image as a Trump loyalist, the harder it will be for her to pivot back to her old persona.
The Road Ahead
Donald Trump and Martha McSally are not doomed to crash and burn in Arizona come November. It’s only June. Many Senators have come back from large polling deficits. Trump has seen his national numbers turn around quickly before. But McSally especially does not seem comfortable with her new political attitude. Her website’s featured ad makes no mention of Trump. It talks of her bipartisan commitment to lower prescription drug prices, her experience as a survivor of sexual assault, and her commitment to Arizonans with pre-existing conditions. The images on the website feature her golden retriever, not the President. Even at the President’s rally McSally seemed uncomfortable, trying to mimic his applause lines but not being able to capture his mannerisms in the same way.
McSally is far from a terrible candidate. In a different time and place she may very well have been a good fit for Arizona. Her career in the House put her far more in line with the John McCains and Jeff Flakes of the world than it did with the Donald Trumps. If 2018 was the sixth year of the Romney Administration, maybe she wins. But in the real world, Donald Trump, and not Mitt Romney, is in the White House. McSally is going to have to win in 2020 with the hand she’s been dealt, and it’s not a strong one. Donald Trump is a poor fit for the state and his numbers nationally, as of today, are subpar at best. Democrats landed a top level Senate recruit whose fundraising numbers are so extraordinary some wonder if he’s stashing a money printer in his basement.
Arizona is only 11 electoral votes, but that’s more than Wisconsin has. The GOP has begun to sound the alarm bells on Texas, Georgia, and the whole Sunbelt. It’s time to look seriously at the possibility that Arizona, a state which catapulted Barry Goldwater and John McCain to national prominence, may be the very one that sends Donald Trump home packing.
