With President Trump, it all comes back to November 2016. So we are not surprised when, during a visit to Poland, he uses his electoral vote tallies as a means of establishing his bona fides.

“As you know,” he said during a news conference Thursday morning in Warsaw, “Polish Americans came out in droves, they voted in the last election, and I was very happy with that result.”

“The Poles have not only greatly enriched this region,” he said a bit later at a public event, “but Polish Americans have also greatly enriched the United States. And I was truly proud to have their support in the 2016 election.”

If this weren’t Trump, we’d probably find this assertion a little weird as a means of making his case to the people of a country he was visiting. But since it is Trump, a different question arises: Is what he said true?

Before we answer that question, we’ll note that, in an election in Poland, Trump might not fare that well. The Pew Research Center released data last month showing that, as with a number of other European countries, views about the U.S. president sank in 2017 with the election of Trump though views about Americans remained constant.

George W. Bush was fairly unpopular internationally and Barack Obama popular. Trump’s numbers in Poland are worse than Bush’s low. (This is probably why the Polish government ginned up supportive turnout for Trump’s visit.)

But, again, that’s not the question. The question is how Americans of Polish heritage felt about Trump’s candidacy.

The political data firm L2 assesses the likely heritage of voters using a number of indicators. Per its data, the distribution of Polish voters in the United States looks something like this — colored by party identification.

What stands out is the density of the population in the upper Midwest — particularly in Wisconsin, Pennsylvania and Michigan, where Trump prevailed by the combined 78,000 votes that handed him the electoral college majority.

The partisan distribution of those Polish American voters varies widely by state. In Pennsylvania, nearly half are Democrats; in Wisconsin, a plurality are Republican. In New Jersey, the biggest group is none-of-the-above. Nationally, the split is about even.

But remember: Even in Michigan, those 214,000 voters or so are only 3.3 percent of the total registered voter population in the state. The margin in the state was narrow enough that a swing in the Polish American vote could have made the difference, but, as we’ve noted before, a small swing in any number of places could have, as well.

It’s true that the most heavily Polish counties in the United States tended to vote for Trump, as was noted shortly after Election Day.

But, overall, there’s no correlation between the density of the Polish American population and the 2016 vote. Even in those counties where the population is at least 5 percent Polish American, there’s no correlation.

What’s more, the graph above masks that the total populations of these counties — and the total number of voters therein — varies widely. Thirty percent of a county of 1,000 people means a lot less than 29 percent of a county of 10,000 people. Sherman County, Neb., has a population of about 1,500. Portage County, Mich., has a population of 34,000.

If we scale the dots on the graph above to the population of Polish Americans, the result looks like this.

That giant blue bubble at lower left? Cook County, Ill. Chicago. The big one near the center of the chart is Erie County, Pa.

A poll conducted in 2008 suggests that Polish Americans backed Obama over Sen. John McCain (R-Ariz.) by nearly 10 points that year. There are certainly people willing to argue that they helped turn out the Polish American community on Trump’s behalf (see here and here) and anecdotal evidence that Trump enjoyed more support from Polish Americans. It’s hard to say definitively, though, that Polish Americans heavily preferred Trump, much less that they turned out more heavily than in the past.

Again, though, we know Trump well enough by now to know that a lack of evidence wouldn’t prevent him from making the claim.