It's also interesting to see what it takes to get a Republican victory in 2012. 40% of the Hispanic and Asian vote combined with a mere 12% of Blacks would have done it.
Another interesting thing is how hugely important Florida becomes to post-2020 Dems in the electoral college. If the Dems fall into the 30's with the white vote, the Republicans end up with a huge advantage in the electoral college until Florida and/or Arizona become a Dem-leaning states because they flip most of the Midwest. I wouldn't be surprised if Florida is left of WI, MN, and IA in the 2020's if current coalitions hold.
Alternatively, assuming the GOP improves it's image with minority voters after immigration reform to such an extent that it gets 40% of the Hispanic and Asian vote, the Democrats only need to get up to 42%-43% of the white vote to compensate. So the conservative base risk to the GOP from immigration reform is real.
Meh. Problem is that it assumes changes in both demographic indicators and voting patterns are uniform across the country. Which of course they never are (e.g., black population is surging in Georgia but dropping in the Northeast; whites are becoming more Republican in the South but more Democratic in the Pacific Northwest).
