President Trump is at the most precarious political moment of his presidency. Or at least, the most precarious since the summer and fall of 2017 when, in the wake of Charlottesville, the failure to repeal Obamacare, and escalating tensions with North Korea, the president's approval ratings were mired in the mid-to-high 30s. It was only the success of the tax cut bill at the end of 2017 that brought Trump's approval ratings back into the 40s, where they've remained ever since.  

Today, his overall job approval rating sits at 41 percent. Not as bad as 2017, but certainly a dangerous place to be this close to re-election. Of course, this has been a consistent pattern with this president. Like a hammer which only knows how to bash a nail, Trump has one speed. He has never been interested in broadening his base — only in mobilizing it and growing it by targeting and turning out as many Trump friendly non-voters as possible. In states like Wisconsin, Pennsylvania and Michigan, where non-voters are more likely to be white and working class, the theory is that Trump can win by expanding the pool of Trump partisans, rather than trying to win back (or win over), more traditional and frequent voters.

As such, his ability to win re-election is centered on him being as close in his job approval ratings as his popular vote showing in 2016. The closer he sits to 46-48 percent job approval rating in October, the better chance he has to squeak out another narrow Electoral College win. But, when he gets much below 45 percent, his path to Electoral College victory gets more and more narrow.  

The big question today is if Trump can improve his standing before November. Given the fact that he has no real interest in meeting this moment of unrest and anxiety with anything other than his typical behavior and approach, it's not likely that HE has an ability to change perceptions of how he's handling his job as president. According to a recent CNN poll, the top four issues today for Americans are the economy, health care, the coronavirus pandemic and race relations. Trump trails Biden by double digits on all but the economy (where he leads Biden by five points). To recover, Trump needs the circumstances to change. Instead of race relations and a pandemic dominating the headlines, perhaps this fall sees kids back to school and a third-quarter economic report that is rosier than expected. Maybe there's a Supreme Court nomination fight. Or, a significant Biden misstep or fumble. 

However, even when times were better for Trump, polling suggests he has failed to make much — if any — improvement from his 2016 performance in key battleground states. This is especially relevant given that Trump won six states — Arizona, Florida, Michigan, North Carolina, Pennsylvania and Wisconsin — with less than 50 percent of the vote. Without a significant third-party candidate or candidates on the ballot this year, getting a plurality of the vote will not be good enough to win a state. Biden doesn't need to win all of these states to win. He just needs a combination of three of them to get to 270.  

Lots of folks short-hand the results of the 2016 election by highlighting Trump's margin of victory over Clinton instead of his actual vote share. For example, hearing that Trump carried Iowa by 9 points sounds impressive, until you learn that he did so while taking just 51 percent of the vote. Clinton underperformed Obama's 2012 vote share in more states than Trump over-performed Mitt Romney's share of the vote. And, in 2018, GOP gubernatorial candidates in Ohio, Florida, and Iowa all took mostly the same percent of the vote Trump did in their states two years earlier. In Ohio, for example, Trump took 51.3 percent of the vote; two years later, Mike DeWine took 50.4 percent.  

That's why it's more important than ever to understand if Trump's vote share in 2016 was his ceiling, or whether he has room to grow. 

In looking through the state polling taken this year, Trump has failed to show consistent improvement over his 2016 showing in the states he was unable to hit 50 percent in 2016. To be sure, the quality (and quantity) of polling in these states vary a great deal. Some states, like Pennsylvania, have had few quality polls — especially in the last few months.  

Polling taken in late May/early June has seen Trump's position in Arizona, Michigan, North Carolina and Wisconsin, slide anywhere from 4 to 7 points from his 2016 showing. In other words, he's not only struggling to improve on his 2016 standing, but he's losing ground from his already anemic performance. Moreover, even those states where his vote share was just above 50 percent, like Iowa (51.2 percent) or Ohio (51.3 percent), Trump is struggling. The most recent Des Moines Register Poll put Trump's showing against Biden at 44 percent. And, again, even before COVID or protests began, there were signs that Trump's showing in those states in 2016 were a high-water mark. A Marist poll taken in March showed Trump at just 46 percent of the vote to Biden's 47 percent — a five-point drop from his 2016 showing.  

At this point, Trump is very much an underdog for re-election. As we watch for signs of his political recovery over these next few months, don't forget to pay attention to the vote share and not the margin. If Trump is sitting at 42-45 percent in state polls this fall, it's almost impossible for him to get to 50 percent, especially when we know that undecided voters tend to break against an incumbent.   

More from the Cook Political Report