Saturday, October 18, 2008

Presidential Elections and This Blog

On my last post about the Presidential Election, Barack Obama's Funny, one of my frequent commenters and a good blogger himself, Schultz, left this comment:
Well said burgher jon. I hope you are going to join me and other bloggers in a challenge to call and speak to 100 voters before election day. A lot of us bloggers talk about how bad we want change but more of us need to put some work in so we don't get let down on November 5th!
There is a reason I will not be joining this challenge or following the Presidential Election in any serious way on this blog. Don't get me wrong, I clearly have my opinions about the election, having made not one, not two, but three posts on presidential politics already this month. I'm excited about Barack Obama and I truly hope, for my own and America's sake, that he wins on November FOURTH, but I can't, won't and shouldn't have any impact on that election.

Being a math nerd, after writing that paragraph I thought, just for fun, let's figure out exactly how little influence I have. Let's say that through my writing on this blog and by calling 100 people that I could influence 200 people to vote not for John McCain but for Barack Obama. In order to calculate the chances those 200 make a difference, we need some assumptions.
  • The guys over at FiveThirtyEight developed a VERY interesting statistic called tipping point state. It gives the probability that a given state will decide the presidential election. According to their simulations, their is a 6% chance that the electoral map works out in such a way that PA going one way or the other will decide the election (This post explains the metric and the latest results are on the right side of the window). In an effort to be conservative with probability equations, I always try to be, let's round up and say there is a 10% chance that PA will matter in the Presidential Election.
  • In 2004 5,769,590 people voted in the PA election for president. To once again be conservative with my calculations, I'm going to round that down to 5 million, in spite of the consensus that this will be an election with a high turnout.
  • The latest polling data for PA comes to a trend adjusted average of +13 for Obama (keep scrolling down on FiveThirtyEight). I'll round that down to a 55-45 lead. You know what, that makes this game NO fun. Let's say that the racists skew the polling numbers by 13 points and that 50% of PA voters plan on voting for McCain.
Ok, those are all the assumptions we need... now the math:
  • First, the probability that I would have an impact in Pennsylvania. If there is a 50% chance that any given voter votes for Obama and 5,000,000 people vote there is roughly a 1 in 6,000,000 chance that McCain wins by 200 votes or less.
  • Next, since there is only a 1 in 10 chance (10%) that PA matters I have to multiply that number by .1. We now have a 1 in 60,000,000 chance (based on stated assumptions) that I influence the election.
I'm going to attack the following counter-argument before it even comes up... "So what if there's a low chance that I influence the election, it's REALLY important, so I should do it anyway."

Well, that just depends on how REALLY important it is to me. There's a way to calculate this. If you're a rational (in the mathematical sense of the word) person then you would pay exactly $10 for a 1/10 chance at $100. You shouldn't pay any more then that, if someone offers it to you at less you should always take it. With that in mind, let's assume that I've invested 10 of my hours to blog and call 100 people and that my time is worth $40 per hour. That means I've paid $400 for my lottery ticket. What would the prize (an Obama election victory) have to be worth to me to pay $400 for a 1 in 60,000,000 lottery? $24,000,000,000. That's 24 Trillion.

In Conclusion, I should only participate in Schultz's challenge if I think Obama winning is 24 trillion dollars better then McCain winning. I don't. Consequently, I (and my blog) will be focusing on the presidential election as more of a public interest story then an "issue" and will instead be focusing on local issues, where I believe I can make a difference with this blog and my voice.

1 comment:

Schultz said...

Wow, a little over analysis but I appreciate that my suggestion resulted in you showing off some math skills. All I am asking is this - while watching the news or anything else during reasonable hours (6pm to 9pm at the latest) us bloggers should make some phone calls. That was all I was suggesting. It is pretty easy to do - and occasionally you will even strike up a good conversation with a complete stranger. I typically call during the commercial breaks.