Andrew Tanenbaum's electoral-vote website is a wonderful resource for tracking per-state polling results, and he posts maps showing both Obama-McCain and Clinton-McCain general election matchups.

On the surface, these maps currently appear to favor the idea that Senator Clinton would be the stronger general election candidate, as the electoral vote totals of the states she leads is higher than for Obama right now.

But that's not the whole story - while Clinton is leading in more states than Obama, in seven states, totalling 112 electoral votes, she leads by less than 5%, whereas McCain has just 2 states, totalling just 13 electoral votes, with such a small lead over Clinton.

By contrast, Obama's numbers are firmer than Clinton's: he has just 4 states, with 53 electoral votes, where he leads McCain by under 5%, and McCain's totals are weaker in a matchup against Obama: 6 states, totalling 76 electoral votes, are closer than 5%.

So who is really leading here? The margin of error of state polls (twice the standard deviation) is typically 4%, so I created a program to simulate the election using Tanenbaum's polling data. I consider each state's polling error to be independent of each other state's, and then I ran 10,000 trials for each matchup.

The results:
Obama  wins 55.9%, averages 277.1 EV
McCain wins 38.8%, averages 260.9 EV
Electoral tie  5.3%

Clinton wins 46.1%, averages 266.5 EV
McCain wins 50.3%, averages 271.5 EV
Electoral tie 3.6%

Clinton is leading McCain, but her support is softer, so if the current polling data are accurate, she actually would be slightly less likely to win in November than McCain. Obama's slight lead over McCain results in a much larger difference in winning percentage.

Remember, this is simply a snapshot of the current state of affairs. I've followed Tanenbaum's methodology of using the most recent polls in each state, averaged with any other polls within a week of the most recent. Some states have more recent polling than others.

What Tanenbaum's maps show, though, is how different the general election would be with Obama running compared to Hillary. If Obama is the nominee, the deep south becomes even more solidly red, but Obama is tied with McCain in North Carolina, and amazingly is just a point back in Texas. An Obama campaign would challenge McCain in many traditionally solid Republican states.

Clinton is weaker in the upper midwest and the Pacific northwest, but she currently has both Ohio and Florida barely in her column, while Obama trails in Ohio and is currently not competitive in Florida. But Obama compensates by putting Texas very much in play (although I'll note the last Texas poll was taken in late February, when Obama was saturating the airwaves prior to the two-step primary/caucus, and also before the "crises" of pastorgate and bittergate).


Winning a state's primary doesn't always mean
you'd do better in the projected general election. Although Obama won
Missouri, McCain would beat him while losing to Hillary. But McCain
leads Clinton fairly comfortably in Michigan, while barely losing to
Obama. So perhaps the Michigan voters don't feel so dissed by the lack
of a revote. Obama loses by less in Indiana than Clinton does, which perhaps suggests he may also be stronger for the primary.


Obviously the dynamics of the race can, and will, change from what we see today. And some of the state polls are a little stale. But based on the most recent state-by-state data, both candidates are in a tight matchup, with Clinton running ever so slightly behind, and Obama running a little stronger.  Either candidate would be electable in the fall.