We’re only three months out from Election Day so you can’t fault them for beginning to think strategically. Every dollar that’s wasted on a state that’s a lost cause is a dollar that can’t be spent in a state that isn’t.

But remember, Trump has a huge fundraising advantage over Biden by dint of his incumbency and his ability to raise money for the general election while Sleepy Joe was still bogged down in primaries. If he’s backing away from a major Rust Belt state in late July, it’s because he’s in a deep enough hole nationally that every last dollar might be needed to successfully perform electoral-college triage. He’s not getting 306 electoral votes this time. It’s unlikely, although not impossible, that he’ll flip any Hillary 2016 states from blue to red. The game in 2020 is simple: Get to 270 by hook or by crook. If that means cutting Michigan loose and spending all of the money allocated for it on must-win Florida and Arizona instead, hey.

He can afford to concede Michigan plus 20 more electoral votes to Biden and still win a second term (coincidentally, Pennsylvania has exactly 20 electoral votes) but it’s a psychological blow to see the campaign temporarily retreat from the site of one of Trump’s most stunning upsets four years ago. Per the Times, they still have $11 million in ads tentatively reserved for September, so maybe they’ll be back. For the moment, though, they’re fighting a defensive action.

Since the end of June, Mr. Trump has spent more money on ads in 10 other states — with Michigan falling behind even much smaller states like Iowa and Nevada — and in recent days, Mr. Trump’s campaign stopped buying ads in Michigan entirely…

Mr. Trump faces a trifecta of troubles in Michigan, according to political strategists and state polling: reduced support among less educated white voters in a contest against Mr. Biden compared with Hillary Clinton; motivated Black voters in the state’s urban centers; and suburban voters who continue to flee Mr. Trump’s divisive brand of politics…

There are uniquely local factors hampering the president, too: Mr. Trump’s unprovoked and unfulfilled threat this spring to “hold up funding” to the state because election officials planned to send absentee ballot applications to voters, as well as his loud sparring with Michigan’s Democratic governor, Gretchen Whitmer, over her response to the coronavirus pandemic. Voters now consistently rate her performance on the issue positively and his unfavorably…

Bradley Beychok, the president of American Bridge, a Democratic super PAC that has made Michigan a major focus, argued that the ad spending disparity was a sign that “Trump is conceding part of the battleground.” If it holds, he said, “this is a big, seismic event.”

The campaign hasn’t bought a TV ad in Michigan since July 3. They’re still organizing in the state and pro-Trump Super PACs are set to start spending there soon in hopes of keeping Biden’s lead manageable in case events this fall give Trump some traction nationally, but it’s easy to see why Team Trump might back away now. Biden leads by 8.0 points in the RCP average, his biggest margin in any battleground state. Within the past week he’s flirted with topping 50 percent on average whereas Trump hasn’t managed to top 42 percent in any poll taken this month. Until this morning’s Change Research survey, which put him ahead by four points, Sleepy Joe hadn’t led by less than six in Michigan since June.

The Trump campaign is probably paying attention to this year’s primary results as well. One of the earliest hints of Clinton’s weakness in 2016 came from the Michigan Democratic primary, when Bernie Sanders pulled the upset of the year — until November 8 — by winning by a point and a half:

Sanders hoped Michigan would deliver for him again this year and blunt the Joementum that Biden had picked up with his string of wins on Super Tuesday. The good news for him is that he got nearly as many votes this year as he did in 2016 (577,000 versus 595,000). The bad news is, uh…

Biden got 838,000 votes and won by nearly 17 points. There’s good reason to believe he’s considerably stronger in Michigan than Hillary was, and even a weak Clinton managed to come within 11,000 votes of beating Trump. Imagine what suckers Team Trump would be if they poured cash into Michigan to make a likely defeat there seem close-ish while Biden is threatening to pull off a momentous, election-ending upset in Ohio across the border.

To make matters worse, Michigan governor and Trump nemesis Gretchen Whitmer has managed to sustain a positive job approval rating even as other governors have seen their numbers slip lately. A survey published earlier this month found that “the average governor has experienced a 10-point decline in approval between April and June.” Whitmer was one of only five whose ratings had actually improved during that period. Likewise, in this morning’s Change Research battleground poll, she tied for the highest approval rating among the six governors tested at 55/45. Trump probably didn’t help his cause in Michigan by singling her out early for her draconian lockdown measures.

Hope springs eternal that the polls in Michigan and beyond are wrong and that “shy Trump voters” are a hidden force that’ll once again deliver an incredible shock on Election Night. Three smart poll analysts coincidentally have analyses of that theory out today and all three are skeptical that hidden Trumpers are being missed in the data. Conservative Henry Olsen argues in WaPo that the shift against Trump among independents lately seems real, and that it explains much of his recent decline. Not only that, Olsen adds, but the “shy Trump” effect four years ago was really more of a “late deciders” effect, as people who didn’t make up their minds until the final few weeks of the race broke heavily for the challenger — as tends to happen in elections. Trump isn’t the challenger this year, he’s the incumbent. He can’t count on that same late windfall.

Meanwhile, HuffPost’s Ariel Edwards-Levy points out that the traditional theories for how Trump voters were supposedly missed last time have already been tested and found wanting:

If support for Trump was uniquely stigmatized, you’d expect him to outperform his polls more than other Republican candidates did theirs. An analysis of 24 live-caller surveys in the AAPOR report, however, found Trump outperformed his estimates by an average 1.4 percentage points in 2016, compared with a virtually identical 1.3 points for GOP Senate candidates

Under the shy Trump voter theory, you’d also expect to see a pattern in how Trump’s numbers were affected by the mode in which a poll was conducted. Specifically, his support should have been consistently lower in polls that used live interviewers (in which Trump voters would need to admit their support out loud to another person) and higher in those conducted online or using automated phone calls (in which Trump voters merely had to click a button). This, however, didn’t happen. Nor was there any evidence of a relationship in the polls between support for Trump and the rates of voters saying they were undecided or were refusing to answer.

It’s possible, said Edwards-Levy, that Trump voters are being honest about their support for him when answering polls but that they’re less likely to answer polls in the first place, which might lead to them being underrepresented in samples. But Nate Cohn of the Times says there’s no evidence for that in this year’s data. If anything, Trumpers might be overrepresented:

But perhaps surprisingly, registered Republicans were actually more likely than registered Democrats to respond to the Times/Siena survey. Over all, telephone calls to registered Republicans or those who participated in a recent Republican primary were about 12 percent likelier to yield a completed interview than calls to Democrats were. This seemingly noteworthy difference can be explained by well-known demographic biases in polling: Older, rural and white voters are likelier than young, urban and nonwhite voters to respond to surveys. After these factors were controlled for, Republicans were no likelier than Democrats to respond to the survey. And if Republicans are just as likely to respond to surveys as Democrats, there’s little reason to believe that they’re vastly underrepresented in political surveys.

When the Times polled six battleground states recently, the members of their sample who voted in 2016 said they favored Trump by a margin of 2.5 points — which is *greater* than Trump’s actual margin in those states four years ago. If the polls are missing “hidden Trumpers,” shouldn’t the 2016 margin among that sample have skewed towards Clinton?

I hope for the GOP’s sake that Trumpers are being undercounted because it’s starting to look hairy downballot too.