Wednesday, February 06, 2008
Super Tuesday: No Decision
Super Tuesday has come and gone. There were some surprises. Huckabee did unusually well, taking five southeastern states. McCain took New York, New Jersey, and California, and that's a lot of big states. He now has the lead, and I predict that he will win the Republican nomination. The real question is whether he can hold onto Lichtman Key 2. That key states "There is no serious contest for the Republican party nomination." This begs the question: What is a serious contest? The answer is anything that amounts to 1/3 of the vote against the nominee. In other words, the nominee must win on the first ballot, and must take at least 2/3 of the vote to secure Key 2. If all the opposition together adds to more than 1/3, Key 2 falls. Delegate vote totals vary widely, but I am getting something like McCain 559, Romney 265, Huckabee 169, and Paul 16 (CNN). The opposition has 450 delegates, and double that exceeds by a huge margin McCain's 559 delegates. It therefore looks like Key 2 will fall. But that isn't certain. What if Romney drops out? His support will scatter among the other candidates, and that may push McCain's total above 2/3. So this key fall is not certain. I had earlier predicted this key would stand, as I had predicted that a single candidate will run away with it. That looks like McCain. But I would now put Key 2 in jeopardy.
The Democrats did not decide anything, either. Obama's late-night Missouri victory may be an indicator of what's to come. It's the bellwether state. (Talk about bell weather. They had some real ding-dong weather last night in the South, and that may have affected returns in Arkansas and Tennessee.) Obama won more states, but Clinton won the big states. It seems to me that Obama wins the "red" states and Clinton wins the "blue" states, although Obama says there are no such things as these types of states. The exceptions to this are Illinois (Obama is a Senator from this state), Connecticut, New Jersey, Tennessee, and Arizona. Clinton has a slight lead in delegates. This may not seem worrisome, but it is like a chess game in which one side is a pawn ahead. By the end of the game, that player could have two Queens. Obama must win more if he is to catch up with Clinton and win the nomination. Virginia will be an interesting state to watch - and that is my state.
How about my exit-poll prediction system? For the most part, it worked. It correctly predicted Huckabee in Georgia as early as 7:15; the networks did not call it until near midnight. It predicted Obama in Alabama and Georgia, and in most of the rocky mountain and plains states. Where it went wrong was with Massachusetts, where it indicated a close race. Instead, Clinton easily won it. It predicted a close race for the Republicans in California, and instead, McCain won it easily. It predicted Clinton for Missouri by 4 points, and suggested that votes for uncommitted and Edwards took the state from Obama. Not so. Obama just barely won the state early this morning. Why did the exit-poll method fail in these instances? Because it calculates the overall exit-poll vote for the candidates. This compares with exit polls predicting candidates for the November elections, and remember what happened with Florida in 2000. For Gore? Well, no there is a problem. And then for Bush? The method isn't perfect. But it has a better record than polls, certainly.
The Democrats did not decide anything, either. Obama's late-night Missouri victory may be an indicator of what's to come. It's the bellwether state. (Talk about bell weather. They had some real ding-dong weather last night in the South, and that may have affected returns in Arkansas and Tennessee.) Obama won more states, but Clinton won the big states. It seems to me that Obama wins the "red" states and Clinton wins the "blue" states, although Obama says there are no such things as these types of states. The exceptions to this are Illinois (Obama is a Senator from this state), Connecticut, New Jersey, Tennessee, and Arizona. Clinton has a slight lead in delegates. This may not seem worrisome, but it is like a chess game in which one side is a pawn ahead. By the end of the game, that player could have two Queens. Obama must win more if he is to catch up with Clinton and win the nomination. Virginia will be an interesting state to watch - and that is my state.
How about my exit-poll prediction system? For the most part, it worked. It correctly predicted Huckabee in Georgia as early as 7:15; the networks did not call it until near midnight. It predicted Obama in Alabama and Georgia, and in most of the rocky mountain and plains states. Where it went wrong was with Massachusetts, where it indicated a close race. Instead, Clinton easily won it. It predicted a close race for the Republicans in California, and instead, McCain won it easily. It predicted Clinton for Missouri by 4 points, and suggested that votes for uncommitted and Edwards took the state from Obama. Not so. Obama just barely won the state early this morning. Why did the exit-poll method fail in these instances? Because it calculates the overall exit-poll vote for the candidates. This compares with exit polls predicting candidates for the November elections, and remember what happened with Florida in 2000. For Gore? Well, no there is a problem. And then for Bush? The method isn't perfect. But it has a better record than polls, certainly.
