Donald Trump holds a dominant position in national polls in no small part because he is extremely strong among people on the periphery of the Republican coalition.

He is strongest among Republicans who are less affluent, less educated and less likely to turn out to vote. His very best voters are self-identified Republicans who are nonetheless registered as Democrats. It’s a coalition that’s concentrated in the South, Appalachia and the industrial North, according to data provided by Civis Analytics, a Democratic data firm.

Trump’s huge advantage among these groups in fact poses a challenge for his campaign, because it may not have the turnout operation necessary to mobilize irregular voters.

It also poses a challenge for the Republican Party, which has maintained its competitiveness in spite of losses among nonwhite and young voters by adding older and white voters, many from the South. These gains have helped the party retake the House, the Senate and many state governments. But these same voters may now be making it harder for the party to broaden its appeal to nonwhite and younger people — perhaps even by helping to nominate Trump.

The Civis estimates are based on interviews with more than 11,000 Republican-leaning respondents since August. The large sample, combined with statistical modeling techniques, presents the most detailed examination yet of the contours of Trump’s unusual coalition.

The estimates reflect the race as it was on Dec. 21, when Trump led a surging Ted Cruz, 33 percent to 20 percent. The modeled estimates do not include people who are undecided, so all of the tallies are modestly higher than they would be if the survey were reporting the unmodeled results with undecided voters.

Perhaps above all else, the data shows that Trump has broad support, spanning all major demographic groups. He leads among Republican women and among people in well-educated and affluent areas. He even holds a nominal lead among Republican respondents that Civis estimated are Hispanic, based on their names and where they live.

But Trump’s lead is not equal among all groups, or across all parts of the country. His support follows a clear geographical pattern. He fares best in a broad swath of the country stretching from the Gulf Coast, up the spine of the Appalachian Mountains, to upstate New York.

Trump’s best state is West Virginia, followed by New York. Eight of Trump’s 10 best congressional districts are in New York, including several on Long Island. North Carolina, Alabama, Mississippi, Tennessee, Louisiana and South Carolina follow.

His strength in the South is blunted only by Ted Cruz in Texas and Mike Huckabee in Arkansas. (Huckabee, despite his weakness nationally, still holds a lead in the congressional district of his Arkansas hometown.) Trump fares well in Florida, despite the presence of two candidates from the state.

The margin of uncertainty around the congressional district estimates is plus or minus 8.7 percentage points, even after more than 11,000 interviews and the benefit of modeling. The data also reflects the preferences of Republican 2016 general election voters — a smaller group than all registered voters, but larger than a primary or caucus electorate. But the broad pattern in the Civis data is still clear.

Trump’s strength fades as one heads west. Nearly all of his weakest states — 16 of his worst 19 — lie west of the Mississippi. Trump’s struggles in Iowa might not just reflect a challenge specific to the state; it may simply be the only state from the Great Plains or Mountain West where public pollsters frequently conduct public opinion surveys. His worst is Utah, a traditionally Republican and affluent state.

His geographic pattern of support is not just about demographics — educational attainment, for example. It is not necessarily the typical pattern for a populist, either. It’s almost the exact opposite of Ross Perot’s support in 1992, which was strongest in the West and New England, and weakest in the South and industrial North.

But it is still a familiar pattern. It is similar to a map of the tendency toward racism by region, according to measures like the prevalence of Google searches for racial slurs and racist jokes, or scores on implicit association tests.

“This type of animus towards African-Americans is far more common than just about anyone would have guessed,” said Seth Stephens-Davidowitz, the economist who first used Google search data to measure racial animus and argued that Barack Obama lost 4 percentage points because of racial animus. He is now a contributing op-ed writer at The New York Times.

Racially charged searches take place everywhere — they are about as common as searches for “The Daily Show” or the Los Angeles Lakers. But they are more common in some parts of the country than others.

That Trump’s support is strong in similar areas does not prove that most or even many of his supporters are motivated by racial animus. But it is consistent with the possibility that at least some are. The same areas where racial animus is highest in the Google data are also less educated and older areas, and Trump tends to fare better among those groups — although the effect of Google data remains just as strong after controlling for these other factors.

These areas also include many of the places where Democrats have lost the most ground over the last half-century, and where Hillary Clinton tended to fare best among white voters in her contest against Obama in the 2008 Democratic primaries.

In many of these areas, a large number of traditionally Democratic voters have long supported Republicans in presidential elections. Even now, Democrats have more registered voters than Republicans do in states like West Virginia and Kentucky, which have been easily carried by Republicans in every presidential contest of this century. As recently as a few years ago, Democrats still had a big advantage in partisan self-identification in the same states.

But during the Obama era, many of these voters have abandoned the Democrats. Many Democrats may now even identify as Republicans, or as independents who lean Republican, when asked by pollsters — a choice that means they’re included in a national Republican primary survey, whether they remain registered as Democrats or not.

Trump appears to hold his greatest strength among people like these — registered Democrats who identify as Republican leaners — with 43 percent of their support, according to the Civis data. Similarly, many of Trump’s best states are those with a long tradition of Democrats who vote Republican in presidential elections, like West Virginia.

Trump’s strength among traditionally Democratic voters could pose some problems for his campaign. Many states bar voters registered with the other party from participating in partisan primaries. Other states go further, not allowing unaffiliated voters to vote in a primary; in the GOP race, for example, that would mean restricting the electorate to those registered as Republicans — one of Trump’s weakest groups. This group of states includes many favorable to Trump, like Florida, Pennsylvania and New York.

Another challenge for Trump is that he commands the support of many people considered unlikely to vote. Civis found him winning 40 percent of the vote among those it gave less than a 20 percent chance of participating in the general election, let alone the primary. He held 29 percent among those who had greater than an 80 percent chance of voting in the November election.

Trump’s campaign will need to mobilize these less-likely voters to maximize his strength. But the challenge shouldn’t be overstated, either. Registered Democrats make up just 8 percent of self-identified Republicans in the states with party registration, according to the Civis data. And Trump still leads comfortably among higher-turnout voters and registered Republicans.

Ultimately, his coalition may augur a bigger challenge for the Republican Party than it does for his own candidacy.

It has been argued that Trump’s divisive language may make it harder for the party to broaden its appeal. But the GOP’s increasing reliance on older and less educated white voters, often from the South, made this challenging long before Trump mounted a campaign. Over the long run, the party will need to figure out a way to satisfy its newest converts while maintaining a message that’s appealing to the rest of the country.