Just shy of nine months after the presidential election that made him president, Donald Trump on Friday morning retweeted a map of the results.

That map comes from Ryne Rohla, who took it upon himself to gather precinct-level data from every possible place he could to make a map that’s more accurate than the state-results map with which we’re so familiar.

So while most maps look something like this (from The Post’s election-night coverage) …

… Rohla’s offers a greater level of geographical precision. Those big blue states on the West Coast are actually mostly blue bastions directly adjacent to the water, with a lot of inland red. And, to a lesser extent, the red states are often speckled with blue cities.

When it comes to maximizing the amount of red possible on the map, Rohla’s map is hard to beat. Given that this is Trump’s goal — remember when he passed out electoral maps to Reuters reporters, noting that “the red is obviously us”? — his retweet is not at all surprising.

But, as always, Rohla’s perfectly accurate map still doesn’t do a great job of depicting what actually happened during the election.

Precincts, just like counties, vary widely in their size. Some precincts have a few dozen voters, some have thousands. So Rohla’s map, like the state-level map, obscures the true extent to which Trump won. Especially since, when it comes to the national totals that this map is meant to depict, Trump actually lost by more than 2 million votes.

If we scale county results by the number of votes cast, a different picture emerges. It can be hard to show all of the counties, since they overlap; we’ve randomized them below. But: A much more even mix of red and blue.

We’ve called out three counties in particular.

  • Los Angeles County, where more votes were cast than anywhere else and which backed Clinton by 48 points.
  • D.C., where Clinton got her highest vote margin.
  • Roberts County, Texas, where Trump got his.

In Clinton’s best county, more than 280,000 votes were cast. In Trump’s, 550 were. D.C. is about 68 square miles in size. Roberts County is 924 square miles.

Here are those two factors, shown to scale.

Trump’s map — and the normal electoral map — show the blue data. Ours shows the green data, which more impartial observers would argue is a more accurate representation of the actual will of the people.

This is not a new clarification, of course. Every since the heady red-versus-blue debate began, arguments over maps that focus on land area have been common.

But it seems that the point still hasn’t really sunk in for those who would like to constantly revisit the results of November’s election. Since we know that everyone is committed to accurately representing what happened nine months ago, consider this our contribution to improving the debate.

Donald Trump won the presidential election. Yet, since Nov. 8, Trump has tweeted about Democratic rival Hillary Clinton many times. (Victoria Walker/The Washington Post)