Washington CNN  — 

First things first: The theme song of the week is the theme from The Larry Sanders Show.

Poll of the week: A new NBC News/Wall Street Journal poll finds that President Donald Trump’s approval rating among voters nationwide is 45%. His disapproval rating is 52%. This makes for a net approval (approval - disapproval) of -7 points.

The average poll has Trump’s net approval among voters at -9 points.

What’s the point: Trump’s 2020 strategy seemed to be in full swing this past week. He looks to be doubling down on a campaign of hardline immigration policies and racial resentment. Trump lashed out in a racist Twitter tirade, telling four ” ‘Progressive’ Democrat Congresswomen [to…] go back and help fix the totally broken and crime infested places from which they came.”

The Trump strategy is pretty simple: 1. Drive up the unfavorable ratings of his Democratic rival as he did in 2016 in order to compensate for his own low ratings. 2. Bank on an electoral college/popular vote split as he did in 2016. 3. Use a campaign of racial resentment to drive up turnout even more among groups favorable toward the President.

Remember, Trump lost the popular vote in 2016. He won because he flipped a congressional district in Maine (good for one electoral vote) and the states of Florida, Iowa, Michigan, Ohio, Pennsylvania and Wisconsin from blue in 2012 to red in 2016. If the 2020 Democratic presidential candidate wins all the states Hillary Clinton won, then she or he only needs to win the closest of those three Trump flipped (Michigan, Pennsylvania and Wisconsin). It seems simple enough.

The potential problem for the Democratic candidate lies in Wisconsin. Trump’s net approval in that state was -4 points in the 2018 exit polls. That was 5 points higher than it was nationally. Taking into account uncontested races, the Democratic House candidates cumulatively won the House vote by 4 points less in Wisconsin than they did nationally.

In other words, the potentially pivotal state in the electoral college may be 4 to 5 points more Republican than the nation as a whole in 2020. This means the opportunity for an electoral college/popular vote split in Trump’s favor remains quite plausible.

Still, Trump would lose Wisconsin and the presidential election if the same people came out and voted for the same party in 2020 as they did in 2018.

This is where Trump’s strategy of going after four very progressive congresswomen comes into play. Trump wants to recreate the 2016 dynamic of making the eventual 2020 Democratic presidential nominee unpopular. Trump would love to tie Reps. Alexandria Ocasio-Cortez of New York, lhan Omar of Minnesota, Ayanna Pressley of Massachusetts and Rashida Tlaib of Michigan to whomever wins the Democratic nomination for president.

Trump’s making a bet that he can link the Democratic nominee to this “squad” – the aforementioned group of congresswomen of color that he has lobbed attacks at. Remember, that’s exactly how he won in 2016: an unpopular Democratic opponent.

In Wisconsin, for example, Trump had a favorable rating of only 35%. He won the state because he ran up a 37-point margin against Clinton among the 22% of the electorate who had an unfavorable view of both Clinton and Trump.

Today, Trump’s less unpopular, but still underwater. To give himself a shot, Trump would be wise to try to make the Democratic nominee a liberal out-of-touch candidate. Although the effect may not be huge, polling and studies of past elections show a very liberal candidate is probably less electable than a candidate in the center.

As I noted last week, three of the four top candidates on the Democratic side (Sen. Kamala Harris of California, Sen. Bernie Sanders of Vermont and Sen. Elizabeth Warren of Massachusetts) are far more liberal than recent Democratic presidential nominees.

Trump’s also hoping that his methods will have the added potential benefit of helping him in the turnout department in Wisconsin. According to analysis by the New York Times’ Nate Cohn, whites without a college degree made up a much larger percentage of voters who didn’t cast a ballot in 2018 than those who did in the northern swing states like Wisconsin. Whites without a college degree make up a core Trump constituency.

Still, we should realize that Trump’s really trying to thread the needle here. A lot of things can go wrong.

His net approval rating remains negative (currently -9 points), as it has throughout a presidency in which he has focused on hardline immigration policies and racial resentment.

Further, Trump’s Republican Party lost the 2018 midterm elections, which were held under similar conditions as today. Trump’s net approval rating was -9 points in the 2018 exit poll. The result was that Trump’s Republican Party lost 40 House seats and the House popular vote by 9 points.

In the final weeks of that cycle, Republicans were not helped by Trump continuously pointing out in the final weeks of that campaign that migrants were coming up through Central America. Voters who decided in the final month of the 2018 campaign were as likely to vote Democratic as those who decided before then.

It remains unclear if Trump can compensate his unpopularity by demonizing the Democratic nominee this time around.

And remember, Trump also risks raising turnout among nonwhite voters. That’s not a big deal in a state like Wisconsin, which is very white. It could, however, take Sunbelt states moving to the left, like Arizona and Texas, and put them into play for the Democrats.

If either Arizona or Texas go blue, it opens up a lot of electoral college paths for the eventual Democratic nominee.