The political junkies among my readers may enjoy the game of guessing the electoral vote for the upcoming election and might therefore wonder what my guesses would be. I have had an organized outlet for such speculation since ... when was the first one? 1992? ... because my friend BRUX runs a little betting pool every four years. The details of the contest vary a little with each election, but the basic idea is that you guess who will win each state, and you score points equal to the electoral vote totals for each state you guess correctly. All participants submit and entry fee an the winner collects the pot.

Even if this weren't a fun game, I'd still have a financial incentive to play. Bruce doesn't keep a cut of the prize money, so if it were a pure lottery it would be an even bet. Besides that, about half the contestants are Bruce's friends who do it just for kicks but don't really watch the polls all that carefully, and then among the others who do I rank above the mean in terms of attentiveness. Beyond that, sure, it's a guessing game with my fellow political junkies, but each of us goes in with an expectation of roughly 2:1 to 3:1 return on our money.

I have yet to win, but one year (1992, I think) I placed a close second (to Brad Wilson). My memory is dim, and I'm fuzzy on the details. One of us got 49 states right and the other got 48, but Brad beat me by one point. I think whichever of us got 49 states missed GA, but I can't remember if that was him or me. I think the other one missed SD and something else.

After experimenting with more elaborate rules last time, Brux has returned to simple rules again this year — only one entry per person, and a single deadline (Sept 30). Entry was by invitation only, which kept the pool down to just 21 entries. The fee was $5, so the winner gets $105.

My guesses are looking pretty good so far. As of today, I'd estimate my chance of winning it all to be somewhere around 1 in 4. I made what at the time seemed like an aggressively pro-Obama prediction but now seem slightly timid if anything. Most of us picked Obama to win (five have McCain ahead, and one has a tie at 270-270) but mostly by modest margins. My guesses give him 353 electoral votes, and only three others in the contest gave him more.

For those who wonder, I gave Obama all the usual swing states (including OH, FL, VA, CO, and NV), I did give him NC, but I didn't give him MO or IN. If you're an election junkie, that should tell you all you need to know, and if not ... well, you're probably not following this in detail anyway.

If I had it to guess again, the only one I'd change is that I'd give MO to Obama now. At least for the sake of pure guessing I would. For the sake of the contest, if I were allowed to change my picks and leave others the same, there's plenty of room for strategic betting. As it stands, even though I like my picks, there are others nearby who made picks almost the same, crowding me in the choicest territory. To maximize one's chances in the contest, it would probably be more strategic to predict an Obama landslide somewhere in the vicinity of 400 electoral votes. There's hardly any competition out there, and it would be easy to craft a strategic set of picks that would claim pretty much the entire territory. (But then, if I'm imagining I get to change my picks, then why shouldn't everyone else be allowed to change theirs as well?)

Having analyzed the competition, I've determined which states I need to watch on election night. The two states I feel least sure about (MO and NC) aren't likely to affect whether I win the contest, since most of my competitors guess them similarly. On the other hand, there is one guy who matches me exactly on every state but IN and LA, which he gives to Obama. That means if Obama wins IN, I can't possibly win the contest, because he'd be 11 points up on me for IN; LA is worth only 9, and I can't improve my score against him on any other state where we match. A similar logic applies to FL on the other side.

My checklist, then, looks like this:

1. If Obama wins IN, I lose.

2. If McCain wins FL, I lose.

3. If Obama wins MO and McCain wins NC, I lose.

4. If McCain wins OH but not FL, then Obama must win NC and McCain must win MO, or else I lose.

I could extend the list with other mathematically possible scenarios where I also lose, but they are increasingly improbable. If McCain somehow comes back and wins the election, then I'll surely lose the contest, but he'll do that by winning FL. Similarly, if Obama wins in a 400-vote landslide, it'll include taking IN.

10:10:10 PM  [permalink]  comment []