I was planning to post this in the wee hours of election day, but -- barring a catastrophe or news with a significant bearing on the election -- I think this is how it will go:
Bush will win 51 percent of the two-party popular vote.* He will take at least 279 and perhaps well more than 300 electoral votes. The range of uncertainty about electoral votes (EVs) reflects the apparent closeness of the race in many states.
Kerry faces likely-to-certain victory in the District of Columbia (3 EVs) and these 20 States: California (55 EVs), Connecticut (7), Delaware (3), Hawaii (4), Iowa (7), Illinois (21), Maine (4), Maryland (10), Massachusetts (12), Michigan (17), Minnesota (10), New Hampshire (4), New Jersey (15), New York (31), Oregon (7), Pennsylvania (21), Rhode Island (4), Vermont (3), Washington (11), and Wisconsin (10). That's a total of 259 EVs for Kerry. The other 30 States, which are leaning-to-solid for Bush, have 279 EVs.
My method of estimating EVs as a function of popular-vote share indicates that Bush's 51 percent of the two-party vote could yield as many as 318-358 EVs. (Go here and see method 3.) Such a result is possible if Bush takes the 279 EVs of all 30 leaning-to-solid States, then picks off Iowa (7) and Wisconsin (10) -- where the races are tight -- plus a combination of Hawaii (4), Michigan (17), Minnesota (10), New Hampshire (4), and Pennsylvania (21) -- for a maximum of 352 EVs. I think that's the best Bush can hope for, unless his popular-vote share is 52 percent or greater.
All of which assumes, of course, that Bush will take Florida (27 EVs) and Ohio (20). If Bush loses either, we may be in for a long, long night -- and perhaps a long several weeks of recounts and court battles. If -- in the end -- Bush loses both Florida and Ohio, Kerry wins.
Republicans are poised to pick up six Democrat seats: Florida, Georgia, Louisiana, North Carolina, South Carolina, and South Dakota. Democrats will probably pick up three Republican seats: Alaska, Colorado, and Illinois. That's a net GOP gain of three seats, for a 54-46 advantage in the Senate.
The best the Republicans can hope for is 55-45, with a come-from-behind win by Republican Murkowski in Alaska. However, that gain could be canceled by a come-from-behind win by Democrat Daschle in South Dakota.
* I base my estimate of Bush's popular-vote share on the presidential vote-share market at Iowa Electronic Markets, Rasmussen's presidential tracking poll, and the "poll of polls" at RealClear Politics.