Keep in mind, though, that while certain voting blocs are considered crucial to winning an election, that may or may not be the case. Depending on your source, blacks only constitute 10-15% of the entire population of the US. Currently, Hispanics are about the same, though a lot of studies suggest that the Hispanic population of the US might grow significantly within the next handful of election cycles, which will make them important. If you take the average values, those two groups together make up only about 30% of the population. Now, I do realize that a bloc can wield influence disproportionate to its size, if that bloc turns out heavily to vote and another bloc does not: black's 15% can mean more if 95% of them vote in an election compared to, say, someone who has 30% of the population, but only 25% of them turn out to vote.
A couple podcasts I listen to (with a grain of salt, as always) have analyzed this election versus the past election. In one of those analyses, they talked about the importance of being able to carry certain groups - Romney vs Obama as the example. The host went on to say that people said Romney didn't court the Hispanic vote enough, he should have done more, etc etc. The numbers analysis of the election showed that for Romney to win, he would have needed (I'll have to fact-check myself later) somewhere around three-quarters of the Hispanic vote (when his polling numbers showed him in the 20% range), whereas he could have accomplished that with only another ~5% of the white vote.
That critical single-digit percent didn't happen, however, because he failed to successfully garner white working-class people, because of attack ads like the one where the guy get fired from his job (because of Romney), lost his insurance, and then his wife died of cancer - or so the story said.
That all said - blacks have characteristically been Democratic voters since the civil rights era, if Trump can pry blacks away from the Democratic party, that'll be a feat unto itself.