Current Standings; Electoral College gives Bush advantage

Kerry has a 53% chance of winning the election, with 275.0 expected electoral votes; however, if the election were held today, Kerry would have a 56% chance of winning, with 275.2 expected electoral votes. Kerry gets an expected 50.1% of the popular vote. This swing comes from a whole bunch of new swing-state polls released today, and also from two small technical changes: I increased the assumed correlation between opinion shifts in different states after I studied the results of some simulations and found quite a few results that violated common sense (for example, Kerry winning Virginia but not West Virginia, or Kerry winning Arkansas but losing the election, or Kerry winning the popular vote my several percent but losing the election); I also increased the assumed daily variance for each state's opinion, so that the variance in the national popular vote would be the desired 0.000001. These new parameters seem to make sense in simulation.

Bush seems to have a half-percent advantage in the Electoral College; this advantage does depends weakly if at all on today's changes in parameters. By this I mean that the Bush and Kerry would have equal chances of winning if Kerry got 50.5% of the two-party popular vote. If the popular vote were 50.0%, Bush would have about an 80% chance of winning. Bush is 9.5 times more likely than Kerry to win the Electoral College while losing the popular vote; indeed, if Bush wins the Electoral College, there is a 36% chance that he will do so without winning the popular vote. Part of the reason for this advantage is that Kerry "wastes" so many popular votes, particularly in California and New York. Another reason is that the small states, which generally support Bush, have more electoral votes per capita than the large states. Another reason is that former secessionist states, all of which support Bush, tend to have lower voter turnout. Another reason is the fairly high likelihood (around 1.2%) of a tie in the Electoral College, which I count as a Bush win. All these things make up for Bush's "wasted" votes in the small Western states.

Because of the increased correlation between the states, even small swing states can serve as good indicators of the national winner; thus, the top swing states are now West Virginia, Virginia (more polls needed, pollsters!), and Wisconsin, followed by Ohio, Pennsylvania, Florida, Michigan, Missouria, Arkansas, Minnesota, and Iowa. Your vote matters the most in West Virginia, Arkansas, and Nevada, followed by New Hampshire, Missouri, Wisconsin, Virginia, Arizona, Colorado, Ohio, and North Carolina.

Cartogram showing, for each electoral vote, the chances that the leading candidate would win if the election were today:

Cartogram showing, for each electoral vote, the chances that the leading candidate will win on Election Day:

Cartogram showing, for each electoral vote, the expected percentage (of Bush/Kerry votes only) for the leading candidate.

Extremely Safe (99%+ chance of winning)

Kerry 146: California, Connecticut, District of Columbia, Illinois, Maryland, Massachusetts, New York, Rhode Island, Vermont.

Bush 125: Alabama, Alaska, Idaho, Indiana, Kansas, Kentucky, Louisiana, Mississippi, Montana, Nebraska (3rd), Oklahoma, South Dakota, Tennessee, Texas, Utah, Wyoming.

Very Safe (90-99%)

Kerry 35: Hawaii, New Jersey, New Mexico, Washington.

Bush 28: Georgia, Nebraska (at-large, 1st, 2nd), North Dakota, South Carolina.

Safe (80-90%)

Kerry 30: Delaware, Iowa, Maine (at-large, 1st), Minnesota, Oregon.

Bush 34: Arizona, Colorado, North Carolina.

Very Probable (70-80%)

Kerry 42: Michigan, New Hampshire, Pennsylvania.

Bush 32: Florida, Nevada.

Probable (60-70%)

Kerry 1: Maine (2nd).

Bush 30: Arkansas, Missouri, Virginia.

Tossup (55-60%)

Kerry 10: Wisconsin.

Bush 0.

Close Tossup (50-55%)

Kerry 0.

Bush 5: West Virginia.

Total

Kerry 264.

Bush 274.