With the defeat of Hillary Clinton and the election of Donald Trump, Democrats may feel they have hit bottom.

The new power structure in Washington will combine a Republican president and a Republican Congress for the first time since 2006. Throw in pending and prospective vacancies on the Supreme Court, and you can see why many progressives consider this the worst-case scenario.

But it is not.

Consider what would have happened if the split decision last Tuesday had been reversed, with Trump winning the popular vote and Clinton the Electoral College (instead of the other way around). Would Trump have accepted the outcome? Would his core supporters have accepted it? That was the nightmare conjured by Trump's refusal, in the final debate, to commit to accept the Election Day results.

But there is another scenario that was always more likely, and it, too, would augur even harsher circumstances for Democrats in the years ahead.

Let's say Clinton had won both the popular and the electoral votes, and even picked up four Senate seats (instead of just two). She would have taken office with a 50-50 Senate and a hyper-hostile House. Her chances of achieving more of the agenda she shared with her predecessor would have been slim to none.

On health care, immigration, economic policy and social issues, the White House and the Hill would be at loggerheads from the outset. The House would strive mightily to hamstring her with investigations. More importantly, the legislative gridlock of the last several years would almost certainly persist.

Next would come the 2018 midterm elections, as always a referendum on the new president. Rocky as this terrain is usually, it would be especially forbidding because of the 2018 map for both Senate and House.

In two years, Senate Democrats will have to defend two-thirds of the seats on the ballot, half a dozen of them in states Trump won handily. In the House, the districts drawn for Republicans by Republicans after the last Census (and the historic GOP landslide of 2010) should continue to safeguard the House GOP majority.

All that means Clinton would have four years of a veto-pen-and-appointment presidency, much as Obama had in his second term. And even her opportunity to stock the federal judiciary — including the Supreme Court — would be in doubt, trammeled by the Republicans' resolve to resist.

In the last weeks of this fall's campaign, several Republican senators said they would be united against any Clinton appointee to the Supreme Court seat of Antonin Scalia, who died in February. And Republicans had already been holding up the confirmation of other judges further down the food chain.

Would there be a public outcry about such resistance? Yes. Would it matter? Probably not any more than the Republicans' flat refusal to have hearings or a vote on Merrick Garland, Obama's nominee to succeed Scalia. Were there any electoral consequences for that?

All of which brings us to 2020, which would be Clinton's prospective re-election year. Whether she actually sought a second term or not, that election would be about her. Given all the factors , plus whatever reversals might happen in the ensuing four years, would we expect the anger we witnessed in the electorate this year to be gone in 2020?

It would have been feat enough for Clinton to defy the eight-year pendulum swing by winning in 2016. To take that out beyond the "third term" of Obama to a fourth-consecutive Democratic term would be to defy gravity itself. Ask George H.W. Bush about 1992 and what he faced after serving the "third term" of Ronald Reagan.

So we are imagining an uphill struggle for a Clinton re-election, especially given the outlook for Congress and the races in the states. And a defeat in 2020 would be disastrously timed for Democrats, because 2020 is also the date of the next census. The national headcount will launch the next round of redistricting, as the last was launched in 2010. If triumphant in that decennial year, the GOP could look forward to another decade of running downhill in most congressional and legislative elections.

Taken together, all these rather predictable developments would be worse yet for Democrats than the setback they are experiencing now. The alternative, at least in the minds of progressives, would seem to be a potentially troubled and controversial term for Trump.

So stop and think about it. Democrats simply cannot expect to move legislation again until they can regain control of Congress. And all signs are that it will take a Republican president, and voter dissatisfaction with a Republican president, to make the Democrats truly competitive in congressional races again.

As we have said, the day of reckoning is not likely to come in 2018 for structural reasons, including Democrats' pathetic turnout in midterm elections. If voters are to have a change of heart about Trumpism, it is a better bet to come in 2020.

So that builds pressure on 2020, a fortuitously numbered year that could be the next hinge in our political history. That could be an advantageous case of timing for the Democrats, a great year for a comeback for all the reasons it would have been a disastrous time for a punishing rejection.

All of this is mere projection, and it may not ease the pain of a narrow loss in a presidential election. But it paints a realistic picture of what would have come next. And for Democrats, the prospect of losing the presidency in 2020 would clearly be worse.

What Democrats have to do is adjust their thinking and their timeframe. They should stop trying to maintain what they won the last decade (mostly in 2006 and 2008 while George W. Bush was still in the White House) and start thinking about how a Republican president can help them rebuild. They need to go back to the base and raise a new pyramid from the ground up, with a new generation of candidates and activists and motivators. There need to be new approaches to issues, new messages to take to the disaffected.

The Democratic campaign of 2016 came down to relying on two categories: minority voters and Trump's character issues. There were lots of both, to be sure, but not quite enough of either.

The African-American vote for president was down by more than a million from 2012. Give Clinton back those Obama voters — especially in Milwaukee, Detroit and Philadelphia — and she would be the president-elect. But perhaps it should surprise no one that Clinton was not the motivator Obama had been for this community. What is more stunning is that the Hispanic vote, while much enlarged, went nearly 30 percent for Trump, according to the exit polls. (Those numbers are in question, however, as Latino Decisions conducted its own exit poll of Hispanics and found just 18 percent for Trump. Clearly, Latinos were a factor and did turn out in big numbers in places like Texas, Arizona, Nevada and Colorado, and yet weren't reflected as strongly as the results suggest.)

As for the character issues, exit polls showed 60 percent of all voters on Election Day disapproved of Trump. Yet the same polling showed a fifth of those disapproving still voted for him. (That might be enough to explain the difference between the final results and the most misleading polls.)

It is far from clear what the Trump administration will actually be about, let alone what it will accomplish. But it seems certain to abandon or dismantle much of Obama's legacy in policy and programs. Democrats cannot do much about that, having lost their last base of power in Washington and their standing in nearly two-thirds of the state capitals as well.

What they can do is begin the slow process of building back. It was always going to be the next phase. It was only a question of when that phase would begin.

Copyright 2016 NPR. To see more, visit http://www.npr.org/.