For those of us who, during the campaign season, checked in regularly at Nate Silver's 538 blog, it might be useful to cast a backward glance. Everyone knows that, with regard to the presidential race, Silver called all 50 states correctly. But his site featured all kinds of other data and prognostications, including (for example) "tipping point" states--those likely to be the one that puts the winner over the top in the electoral college. When you are watching election returns on television, there is a geographic aspect to the state that clinches it for one or the other contestant, but Silver has in mind something else. In this year's election, won by Obama, you start with the District of Columbia, the venue that he carried by the widest margin (84%). You then add to its 3 electoral votes those of Hawaii (4), which Obama won by 43%, his second widest margin. And you continue in this fashion till you arrive at the state won by Obama that put him past 270 electoral votes.
The tipping point state for the election of 2012 was Colorado. Obama carried it by 51% to 47%, and it moved him from 263 to 272 electoral votes.
Some observations.
1. Obama's electoral vote total of 332--62 more than the minimum needed to win--suggests he had considerable margin for error. The tipping point analysis, however, tells a slightly different story. Colorado plus 22 other states that Obama won by more that 4% gave him what he needed. His remaining 60 electoral votes were supplied by just three (big) states: Virginia, Ohio, and Florida (where his margins were 3%, 2%, and 1%, respectively). So you could say that Obama's margin for error was 62 electoral votes. But it would be more accurate to say that his margin for error was the tipping point state plus just three more, all of which he won either narrowly or very narrowly.
2. You can reverse engineer the analysis. Had Romney been able to win Florida, Ohio, and Virginia--and he was close in all three--then Colorado could have been the tipping point state of a Romney victory.
3. The exercise of ordering the states, from Obama's easiest victory (Hawaii) to his most lopsided defeat (Utah), reveals how uncompetetive a relatively close election can be. A hundred nineteen of Obama's electoral votes were won in seven states (plus the District) where his margin of victory was more than 20 points. For more than a generation now, no Republican presidential candidate has come close to winning the biggest state in the East (New York), the biggest state in the Midwest (Illinois), or the biggest state in the country (California). Looking at it from the Republican side, Romney won 21 states by a double-digit margin. He had single-digit wins in only three states, and two of those-- Georgia, which he won by 8%; and Missouri, which he won by 9.6%--weren't very close, either.
4. Silver, who consistently had Colorado, Virginia, and Ohio as the most likely tipping point states, knows what he's doing. But so do the candidates. The election outcome was determined in Florida, Ohio, Virginia, and Colorado, and those states, together with just a few more, were the ones in which the respective campaigns invested virtually all their time and money.
5. Which is why the electoral college should be abolished. Why should the election be decided by the voters of just a few states? Why should all the rest of us be mere spectators? Why should our election analysis include such terms as "tipping point states"? The electoral college is responsible for all these pathologies, and more. Support--I mean agitate for--the National Popular Vote Movement.
