President Obama and Hillary Rodham Clinton. (Mandel Ngan/Agence France-Presse via Getty Images)

BuzzFeed chief Ben Smith sat down with President Obama on Tuesday. The interview drew lots of attention largely because, well, the president sat down with BuzzFeed. (If you are surprised by that, you haven't followed the Obama administration's media strategy very closely.) But Obama makes an important admission in the interview that gets at a critical question facing Democrats as they try to maintain control of the White House in 2016.

Here the exchange:

Smith: You were elected with this new coalition of young people, people of color, women, and I wonder: Is that a coalition that the next Democratic nominee — Hillary Clinton or not — inherits?

Obama: I don’t think any president inherits a coalition. I think any candidate has to win over people based on what they stand for, what their message is, what their vision is for the future.

The central question facing Clinton, the de facto Democratic presidential nominee in 2016, and the legion of former Obama consultants advising her this time around is whether they can turn the Obama coalition into a Democratic coalition. Obama's skepticism on that question is probably born out of his giving a politic answer. (You don't say things like "Yes, I think I can port all of my voters to Hillary" if you are Obama.) But a deeper dive into the Obama coalition suggests that re-creating Obama's winning coalition isn't as simple as just being the candidate with a "D" after your name on the 2016 ballot.

Let's start by looking at exactly what constitutes the "Obama coalition." To do that, I compared exit polling from 2008 to how Sen. John Kerry and then-Vice President Al Gore performed in 2004 and 2000, respectively. When you do that, three demographic groups jump way out.

* African Americans

Obama won 95 percent of black voters in 2008. Kerry won 88 percent in 2004, and Gore won 90 percent in 2000. That improvement in overall margin was made even more important by the fact that black voters comprised a larger chunk of the overall vote in 2008 (13 percent) than they did in 2004 (11 percent) or 2000 (10 percent).

* Young voters (ages 18 to 29)

Obama's margins among young people as compared to past Democratic nominees are startling. He won 66 percent among those ages 18 to 29 in 2008; Kerry took just 54 percent, while Gore claimed just 48 percent. The youth vote has been consistent as a function of the overall electorate: 18 percent in 2008, and 17 percent in 2004 and 2000.

* Hispanics

Obama won 67 percent of the Hispanic vote in 2008, well above the 56 percent Kerry took in 2004. Gore fit in between those two -- taking 62 percent. Hispanics have been very slowly increasing their share of the electorate -- from 6 percent in 2000, to 7 percent in 2004 and 9 percent in 2008.

One other group performed marginally better for Obama than for Gore or Kerry: women (as Smith mentioned in his question). Obama won 56 percent in 2008 as compared with 51 percent for Kerry in 2004 and 54 percent for Gore in 2000. So, women could be considered the third-and-a-half leg of the Obama coalition stool. (Note: That is a terrible metaphor but you get the point.)

It seems very unlikely that Clinton will be able to replicate Obama's performance among black voters given the historic nature of his candidacy as the first African American presidential nominee. Ditto Obama's massive margins among young voters because, well, Obama's candidacy in 2008 was exciting and cool for many young people. While Clinton's candidacy inspires excitement among some groups -- more on that in a minute -- young people are not, to date, one of them.

Whether Clinton can match or better Obama's performance among Hispanics depends in large part on what Republicans do (or don't do) over the next 15 months on immigration policy and whom they nominate for president. If Republicans nominate, say, Jeb Bush or Marco Rubio, both of whom have called for a kindler, gentler message toward Hispanics from their party, it seems unlikely that Clinton could match Obama's performance. If, on the other hand, Republicans pick Ted Cruz, whose hard-line immigration policies appeal to many people in the Republican base, you could see Clinton equal or even improve on Obama's margin from 2008.

Now, just because Clinton doesn't seem likely to replicate the winning Obama coalition doesn't mean she can't get elected president with a slightly different coalition. Her candidacy virtually ensures that she will be the first female presidential nominee in either party, a bit of history that could well allow her to outperform even Obama's showing among women in 2008. Worth noting there, however, is that Obama's 56 percent in 2008 (and the 55 percent he won among women in 2012) are the two best performances for a Democratic presidential nominee among female voters since at least 1980.

And, even more difficult to calculate but worth considering is whether Clinton will do better -- maybe considerably better -- than Obama among white voters. It is true that Mitt Romney won the white vote by the largest margin -- 20 points -- since Ronald Reagan during his blowout win in 1984. But that factoid should be taken with a grain of salt. Yes, Obama's 39 percent among whites was a low ebb for the party in recent elections. But it's not as though Democrats had been killing it with the white vote pre-Obama. Gore won just 42 percent. Kerry took 41 percent. In fact, Obama circa 2008 did better among white voters -- 43 percent -- than either of them.

Looking at the numbers, it seems clear that Obama was right when he told Smith that the Obama coalition isn't just going to emerge in toto for the next Democratic nominee. It almost certainly won't. And that should worry Clinton and everyone advising her on how to win the White House in November 2016.