Odd Hours

Other than exit polls, what are some indications of how the election is going? Since I have some money on the line tomorrow night, this is a question I have given a fair amount of thought. Since the networks are likely to be significantly more cautious about calling states this time around (nobody wants to issue another Florida retraction!), if the vote is close, it may be late in the night before key states like Ohio and Florida are called. But clues can be gleaned from which other states are called and the timing of the calls. Based on this timeline of election night 2000, I've drawn up a set of guidelines.

Ohio: If PA and MI are called at 5pm (all times are PST), it means Kerry is ahead by more than 5 points in these states, and he will likely win Ohio. If neither PA or MI is called by 6pm, Kerry probably loses Ohio. If WV is called before PA or MI, Kerry probably loses Ohio.

Florida: The networks will be especially cautious about calling the Sunshine State this time around. If FL is called for either candidate by 6pm, when the polls close in the panhandle, it means that candidate has a significant enough lead in FL that he will almost certainly win the election.

New Hampshire: If ME (3 of 4 EV's) is called at 5pm, Kerry probably wins NH. If either NH or ME (CD 2) is called for Kerry by 6pm, Kerry likely wins the other one too.

Iowa and Wisconsin: If MN is called before both MO and AR, Kerry probably wins IA and WI. If one of MN is called after one of MO, AR but before the other, Bush has the edge in IA and WI. If both MO and AR are both called before MN, Kerry probably loses IA and WI.

Southwest: This is the trickiest region, because there are four closely related states (NM, NV, CO, AZ). If it comes to this point, we will likely find ourselves wanting to know whether Bush has a chance in NM, or whether Kerry has a chance in NV. If either AZ or CO is called by 8pm, then Bush has a good shot in NM. If NM is called for Kerry by 8pm, or AZ is not called by 9pm, then Kerry has a good shot in NV.

Bush 284, Kerry 254. Same as 2000, except New Hampshire goes to Kerry and Wisconsin goes to Bush. The big prizes, Florida and Ohio, are within reach if we win the turnout war. But there are enough potential problems in these two states -- voter suppression and electronic voing machines chief among them -- that Bush has the edge. Many people would chose Iowa instead of Wisconsin as the likely Bush pickup in the upper-midwest. However, I think the NRA and the suppression of black voters in Milwaukee will end up putting Wisconsin in the Bush column. Finally, although recent polling puts Bush ahead in New Mexico, I think the increased number of Hispanic voters and popular Democratic governor Bill Richardson will put Kerry on top.

While the presidential election could easily go either way, I'm not very optimistic about our senate chances. There are too many open Democratic seats this cycle, and all the competetive races are in red states, where Bush will have coattails. My predictions:

Alaska: Knowles (D)
Colorado: Salazar (D)
Florida: Martinez (R)
Kentucky: Bunning (R)
Louisiana: Vitter (R) surpasses 50%, avoiding a runoff in December.
North Carolina: Burr (R)
Oklahoma: Coburn (R)
South Carolina: DeMint (R)
South Dakota: Thune (R)

New senate: 54 Republicans, 45 Democrats

The potential big news is Daschle's loss in South Dakota. While this would be a short-term negative because it costs the Democrats a seat, I think it will turn out to be a blessing in the long run. We need a senate leader who doesn't have to sell out Democratic values to get re-elected.