The consequences of shifts in presidential election behavior can be enormous. But it’s worth recalling that the actual shifts themselves can be pretty small in the scheme of things.
Here’s Robert Vanderbei’s map of the 2004 election:

And here’s 2008:

Obviously — not the same. But pretty similar.
November 8th, 2008 at 9:04 am
Yes, Matt. If you measure electoral support over land mass things always look pretty good for Republicans…
November 8th, 2008 at 9:04 am
These pure area maps that are not adjusted for population are utterly misleading and useless. “Not a lot of difference” perhaps because the large unpopulated areas with 50 Republicans and 20 Democrats are dominating the visual field, while the NYC, Chicago, and LA areas are too small to see?
Cranky
November 8th, 2008 at 9:20 am
One possible key to a future Republican resurgence is for the people who live in the Red States to have a lot more kids.
Of course, they made not be able to afford them, and would then have to look to the government for a safety net, which could in turn cause them to vote for Democrats, thereby making them Blue States.
November 8th, 2008 at 9:45 am
Here’s the 2000 population distribution map from Census
http://www.census.gov/geo/www/mapGallery/images/2k_night.jpg
November 8th, 2008 at 9:48 am
Acres don’t vote.
November 8th, 2008 at 10:12 am
The “center” in both 2004 and 2008 includes a lot of people who voted Democrat and a lot of people who voted Republican, the balance just shifted a bit. Die-hard partisans of either party will never really be representative, even if their candidate wins. But if Republicans claimed that the average American in 2004 was totally on board with the Republican agenda, we can’t help but claim the reverse. We can now say that New York City and Chicago are “real America” and Oklahoma and Tennessee are “fake America”.
November 8th, 2008 at 10:16 am
Awhile back I worked it out this way: suppose you take today’s public opinion and classify it as 15% left, 35% left-center, 35% right center, and 15% right. (Since we have a pretty hefty hard right at the moment, and a rather weak left, this would mean calling some people “right center” who are pretty far right in a broader perspective, and some people as “left” who are actually pretty moderate. (I’m just talking about making an even distribution out of the present state.)
Then move 10% of each group one place to the left. The new distribution of today’s political positions would be 13.5% right, 33% right center, 35% left center, and 18.5% left. The right is now 46.5% and the left 53.5%. That’s a hefty difference and in our system it amounts to a clear left majority, even though only 10% of the population shifted at all, and no one shifted more than one place to the left.
This is complicated by the distribution of the population within the states, of course, and by the gerrymandering of the Senate favoring small states.
And it’s complicated further by the organization of Congress. The Republican Congress worked by the “majority of the majority” rule, which means that as few of 51% of 51% (26%) of the Congressmen could control Congress.
And since only about half of people vote, 26% represents 13% or so of the population, and throw in gerrymandering, Congress has been controlled by slightly more than 10% of the American population.
This doesn’t mean that the other 90% all disagree. It just means that they didn’t actually participate in electing the people who controlled Congress. But this does show you how the 30-35% of the population who are hard-core rightwingers succeeded in controlling Congress for most of eight years.
November 8th, 2008 at 10:28 am
Re: One possible key to a future Republican resurgence is for the people who live in the Red States to have a lot more kids.
That’s a pretty long pipeline– 18 years at a minimum, longer really until the kids become likely voters. And it’s also a very leaky pipeline as a lot can happen to divert those children in the course of a generation. My father and step-mother were both Republican voters back in the day. I am decidedly not (and neither is my step-mother these days).
November 8th, 2008 at 10:32 am
Hey, Cranky. Do you (or anyone else) have a link to a map that would graphically weight population distribution and then colorize according to district voting tallies?
j
November 8th, 2008 at 10:39 am
Here you go:
http://www-personal.umich.edu/~mejn/election/2008/
November 8th, 2008 at 11:18 am
The 3-d maps are generally easier to understand that make up for differences in population. http://www.washingtonpost.com/wp-srv/politics/interactives/campaign08/election/uscounties.html
When you look at that map, it becomes apparent why the Repulbicans have such lousy long term prospects.
November 8th, 2008 at 11:30 am
Artistic interpretation of what the morning after may have been like for John McCain and Sarah Palin…
November 8th, 2008 at 11:35 am
It’s particular difficult to see the change on this rendering, partially because of the way the color scale is done. But if you move the slider bar between 2004 and 2008 on the New York Time site:
http://elections.nytimes.com/2008/results/president/map.html
the change looks pretty darn significant.
So the change is what it is, but you can tell a different story about it depending on how you present the data.
November 8th, 2008 at 11:48 am
I just realized that if you click on “Voting Shifts” and “County Leaders” on that same page you can see what the county-wise change really was. Really fascinating.
You can also see the shift from 1996 to 2000 was much more significant that 2004 to 2008.
http://elections.nytimes.com/2008/results/president/map.html
November 8th, 2008 at 12:06 pm
yglesias, what’s your broader reason for repeatedly trying to tell people their vote does matter when you have a lot of smart commenters constantly pointing out how and why they do? what are you trying to accomplish?
November 8th, 2008 at 12:19 pm
What Cranky Observer said. This is just stupid. The map shows that rural counties remained static. The shift was in the suburbs and without population adjustment it’s impossible to see on a map like this.
Taking total votes gained by Dems between 2004 and 2008 and votes lost by Republicans lost shows a swing of 10.3 million ballots, or an 8.45% change in the totals between the two parties. The big county splotches on this map often have fewer than 20,000 total ballots cast and they were not the places that swung between parties. To map the 10.3 million vote shift we need better geographers.
November 8th, 2008 at 1:19 pm
Matt, the difference is quite stark when weighted for population.
November 8th, 2008 at 1:47 pm
The only interesting thing about these two maps, is that it looks like the 2008 election was even more polarizing. Of course, the only thing that means in this case is polarizing by geographical regions. So rural voters hated Obama even more than Kerry. And Obama had even more broad based support among urban voters than Kerry. Not too surprising if true, but also not very helpful in terms of identifying important demographic trends and voting patterns.
November 8th, 2008 at 2:43 pm
Re: The map shows that rural counties remained static.
Not entirely, although it’s easier to see in a stark red-blue map. To give just one example: Chatauqua County NY (where I have friends) went Bush in ‘04, Obama this year. There may be a few Blacks and Hispanics in Jamestown and possibly Fredonia, but otherwise it’s a very white, rather rednecky, county where most people drive pickups and own guns– and where the economy has gone to hell in a handbasket.
November 8th, 2008 at 3:27 pm
The middle purples are almost indistinguishable from one another. Horribly designed map. Nice to look at, terrible for conveying information.
November 8th, 2008 at 3:38 pm
Fine, look at the population-weighted versions. Not much change there either:
2008
2004
November 8th, 2008 at 4:21 pm
Matt, are you sure these are using the same color scale? The 2004 map looks much bluer to me when I tab back and forth, which doesn’t make sense given that we know most of the country was bluer in 2008.
@Archit Shah:
I’d say there is a very noticeable change in the weighted versions.
November 8th, 2008 at 4:50 pm
yeah, it’s a minor point, but increasingly I think routine pushback against this type of map, which will always provide an illusion of Republican dominance, is important. visual info persuades and seduces. maps like these need to go away.
November 8th, 2008 at 5:36 pm
I love the maps, they are basically perfect.
November 9th, 2008 at 2:28 am
Well, I don’t think the maps “provide an illusion of Republican dominance”. Part of the point (both 2004 and 2008) seems to be highlighting the purpleness of America. Matt’s point is well illustrated in that the shifts were not in hundreds of counties suddenly going notably blue but thousands of counties getting a little less red, perhaps imperceptibly so at this scale.
Of course there are other maps for other points, but this one has its place and uses.
November 10th, 2008 at 1:51 am
we can a href=http://www.iwowleveling.com>wow gold
give you the gold as soon as possible, face to face or mail, choose you like.
Reliable
We farm the buy WOW Gold by manual,
February 14th, 2009 at 4:45 am
As the financial crisis grips the entire world. We need to be more informed
of the options and choices that we have. Found this website that has a lot
of information regarding reverse mortgage and other mortgage related issues.
The website is dedicated to Mortgage problems faced by real people and the
answers to it. You will find a lot of questions and the answers to those
questions posted by reall real people so it is very informative.
So see the answers to you mortgage questions at What Is Reverse Mortgage
March 1st, 2009 at 11:00 pm
cialis
It is the coolest site,keep so!
March 11th, 2009 at 9:36 am
It is the coolest site,keep so!
March 22nd, 2009 at 7:37 am
tramadol
It is the coolest site,keep so!
April 2nd, 2009 at 10:23 am
I bookmarked this site. Thank you for good job!
buy cheap viagra
April 8th, 2009 at 4:27 am
I want to say – thank you for this!
viagra
April 16th, 2009 at 5:22 pm
Incredible site!
viagra