Announcing: The JWM 2004 Election Day Prediction Contest!

Category: Election 2004: Here We Go Again

Monday, October 25, 2004

The JWM 2004 Election Day Prediction Contest is open -- follow this link to post your predictions!

It occurred to me today that as we come into the home stretch in this election (and Lord, it's been long enough -- my first post in this category is dated May 12, 2003!), it might be fun to liven things up a bit by giving y'all a chance to demonstrate your all-knowing political wisdom. So, it gives me great pleasure to announce the Just Well Mixed 2004 Election Day Prediction Contest.

Here's how it's gonna go down.

The way to win the contest will be to most accurately predict how the Electoral College vote will be distributed. That's right, just like the Constitution, we couldn't care less about the popular vote; it's all about the Electoral College, baby.

The contest will officially be open for submissions on one day: Friday, October 29. On that day, I will open a post for contestants to post their entries. To enter, just add your prediction as a comment on that post. Predictions entered on other posts will be disregarded. The thread will be only be open for posting for that one day, to prevent anyone from taking advantage of more recent poll data to get an edge.

On November 3, after it is clear how the Electoral College votes are going to be distributed, I will announce the winner. The prize for winning will be a copy of the excellent political computer game President Forever, on me. (Contestants who already have PF or who can't use it for whatever reason will receive the cash value of the prize instead.)

Submissions need at minimum only specify the candidates and how many electoral votes you believe each will receive. For example, BUSH 274, KERRY 251 is a perfectly valid entry. However, I encourage you to share how you calculated your prediction -- why you believe certain battleground states will go the way you think they will, for example. If the winner includes this type of background, either in their comment or as a link to an entry on their own blog, I will throw in a copy of President Forever's sister product, Prime Minister Forever (or the equivalent cash value), to say thanks for going to the extra effort.

Some potential complicating factors, and how I plan to deal with them:

• If there are multiple submissions of the winning entry -- in other words, a tie -- everyone who submitted a winning entry will be asked a tiebreaker question by e-mail within 1 week of Election Day. The winner of the tiebreaker will receive the prize.
• If we don't know on November 3 what the Electoral College vote distribution is because of another 2000-style legal debacle, all bets are off (in more ways than one). Given the increasingly likely scenario of one or more states being tied up in Florida 2000-style knots, I'm now allowing entries that predict that we won't know the winner by November 3 due to incomplete counts in one or more states. However, for such an entry to win, you must submit a complete list of the states you believe will be incomplete as of 12:00 noon, November 3, 2004, and your list must be 100% correct.
• If nobody gets a majority in the Electoral College -- who cares? You can still win the prize even as the Republic crumbles around us. There's nothing stopping you from predicting a (mathematically possible) 269-269 split.

So -- it's free to enter, and you could win some cool software. Why not give it a whirl? (And tell your friends, the more the merrier.)

To help you get started, here are some basic facts and resources that can help you figure out what your predictions should be:

About the Electoral College Process

• There are a total of 538 Electoral College votes available.
• To win the Presidency, a candidate must receive a majority of these votes. A "majority" is defined as 50% of the votes plus one, which for a total of 538 means that a minimum of 270 is required to claim victory.
• Each state receives a number of Electoral College votes equal to the number of Members of Congress (Representatives plus Senators) it has. Because states get seats in the House of Representatives based on their population, this means that states with large populations end up with more Electoral College votes than do states with small populations.
• Other states with smaller populations are nonetheless important because they are "battleground states". These are states where theoretically either candidate could convince a majority of the population to vote for him. These states are important because they are the only places where a candidate can win new Electoral College votes -- getting more votes in a state that already favors his candidacy doesn't impact the College in any way. Battleground states in 2004 include Florida, Ohio, Wisconsin, Iowa, and Missouri.

Tools and Resources

• The Electoral Vote Predictor -- the motherlode. EVP is chock full of polling data from every state that matters in this year's contest, which you can see in graph form or as the raw CSV data.
• PBS Politics 101 Electoral College Map -- this neat little app lets you turn states red and blue interactively, and see how flipping them affects the overall count for each candidate. This is a fast and easy way to work up your overall total once you have figured out the trends in the states.

That's it for now. If you have feedback or questions about the contest, drop me a line before Friday and I'll be happy to consider changes where necessary. And now -- start working on those predictions!

UPDATE: Oscar asks what happens if someone submits more than one entry. Could they "spam" the contest by putting in a whole range of submissions covering all possible outcomes? It's a good question. The answer is, no they couldn't, because I am going to only take the latest comment (as timestamped by Movable Type) -- all earlier ones will be disregarded. So you can change your entry if you want, but don't go whining to me if it turns out your earlier prediction was the right one!