Monday, November 01, 2004
Tomorrow is the Day
Well finally after over 2-3, in fact, 4 years of campaigning, we finally cast our ballots and find out who is going to lead the most powerful nation on Earth for the next 4 years. So the question is, who is going to win?
For a long time I thought that President George W. Bush was going to win. This is because of the Keys of the White House theory of Allan Lichtman. The 13 Keys are a set of True and False statements. If 6 or more of the statements are false, the challenging party wins; otherwise the incumbent party wins. For much of the past few years, only four of the keys were false ("had fallen"). So I called for Bush's reelection. However, I believe that two more keys have fallen, namely the Short Term Economy key (Key 5) and the Foreign and Military Success key (Key 11). This makes a total of six keys, and two weeks ago, I called for Bush's defeat.
However, these keys are highly subjective. For example, Key 11 depends on how Bush is doing with foreign policy. He invaded Iraq, and initially succeeded: The Husseins are gone. However, there has been a steady rain of bad news from Iraq, and these have hammered at that success. The Mission is NOT Accomplished. But is there enough failure out of this to cause the key to fall? It seems to me that there is backward stuff going on here: if Bush wins, then Key 11 has stood. If Kerry wins, then Key 11 has fallen. This makes Key 11, and in fact the key system altogether, useless for predicting this election. Of course, the keys have predicted the election again.
So how about other signs? The polls? Just about all the states are settled. Only about 10-12 battleground states remain. I have been watching the polls the past few days, first towards Kerry, then Bush, then Kerry. The last I heard, two of the metapolls had predicted a 269-269 tie. This means a donnybrook and probably a win for Bush.
Ohio and Florida are the two swingingest states. It seems now that to a certain extent, if Kerry wins either one, he wins the election. If Bush gets them both, he wins. I did a spreadsheet analysis with 10 swing states and 1024 combinations. I came up with these results:
This says that indeed, Kerry will win if he wins one of Ohio and Florida and loses if he loses both. However, Florida is more important to win than Ohio, since it has more electoral votes, and if Bush and Kerry split their votes, the prospect of a 269-269 tie ominates strongly.
So that helps. But still we have to wait for results in those two states to come in. And they could be really close; enough to cause the results to be delayed for weeks as in 2000. The difference this year is that we are playing this one with several parts. There could be several Floridas to deal with this time.
How about other indicators? The Green Bay Packers beat the Washington Redskins. That means that Kerry will win, since the incumbent party has always won if and only if the Redskins won their football game just before the election. I put this in the same category as 12-game winning streaks in baseball. Just because a team has won 12 times in a row does not mean chances are good for a 13th win.
A better indicator may the Dow Jones Average. In each year that the Dow declined by half a percentage point or more, the incumbent party has lost. The Dow so far has lost 0.52 %. Therefore, Kerry will win.
Another one is that Kerry will win if he takes Ohio. No Republican Presidential candidate has won without taking Ohio. This is another winning streak. It can end any time.
I am not offering any predictions now. This is about as tied as they can get. I can see what Slate means when they say "Fit to be Tied".
For a long time I thought that President George W. Bush was going to win. This is because of the Keys of the White House theory of Allan Lichtman. The 13 Keys are a set of True and False statements. If 6 or more of the statements are false, the challenging party wins; otherwise the incumbent party wins. For much of the past few years, only four of the keys were false ("had fallen"). So I called for Bush's reelection. However, I believe that two more keys have fallen, namely the Short Term Economy key (Key 5) and the Foreign and Military Success key (Key 11). This makes a total of six keys, and two weeks ago, I called for Bush's defeat.
However, these keys are highly subjective. For example, Key 11 depends on how Bush is doing with foreign policy. He invaded Iraq, and initially succeeded: The Husseins are gone. However, there has been a steady rain of bad news from Iraq, and these have hammered at that success. The Mission is NOT Accomplished. But is there enough failure out of this to cause the key to fall? It seems to me that there is backward stuff going on here: if Bush wins, then Key 11 has stood. If Kerry wins, then Key 11 has fallen. This makes Key 11, and in fact the key system altogether, useless for predicting this election. Of course, the keys have predicted the election again.
So how about other signs? The polls? Just about all the states are settled. Only about 10-12 battleground states remain. I have been watching the polls the past few days, first towards Kerry, then Bush, then Kerry. The last I heard, two of the metapolls had predicted a 269-269 tie. This means a donnybrook and probably a win for Bush.
Ohio and Florida are the two swingingest states. It seems now that to a certain extent, if Kerry wins either one, he wins the election. If Bush gets them both, he wins. I did a spreadsheet analysis with 10 swing states and 1024 combinations. I came up with these results:
| Scenario | Kerry | Bush | Tie |
| Kerry OH and FL | 98.9% | 0.8% | 0.2% |
| Kerry FL Bush OH | 71.9% | 17.9% | 10.2% |
| Bush FL Kerry OH | 51.0% | 41.2% | 7.8% |
| Bush OH and FL | 1.2% | 98.3% | 0.5% |
| Overall | 55.6% | 39.8% | 4.7% |
This says that indeed, Kerry will win if he wins one of Ohio and Florida and loses if he loses both. However, Florida is more important to win than Ohio, since it has more electoral votes, and if Bush and Kerry split their votes, the prospect of a 269-269 tie ominates strongly.
So that helps. But still we have to wait for results in those two states to come in. And they could be really close; enough to cause the results to be delayed for weeks as in 2000. The difference this year is that we are playing this one with several parts. There could be several Floridas to deal with this time.
How about other indicators? The Green Bay Packers beat the Washington Redskins. That means that Kerry will win, since the incumbent party has always won if and only if the Redskins won their football game just before the election. I put this in the same category as 12-game winning streaks in baseball. Just because a team has won 12 times in a row does not mean chances are good for a 13th win.
A better indicator may the Dow Jones Average. In each year that the Dow declined by half a percentage point or more, the incumbent party has lost. The Dow so far has lost 0.52 %. Therefore, Kerry will win.
Another one is that Kerry will win if he takes Ohio. No Republican Presidential candidate has won without taking Ohio. This is another winning streak. It can end any time.
I am not offering any predictions now. This is about as tied as they can get. I can see what Slate means when they say "Fit to be Tied".
