Friday, November 23, 2012

Newcomb's Decision

This post is partially a continuation of my previous posts on utilitarianism, and partially on philosophy in general; mostly, it's my two cents on one of the odder parts of consequentialist debate: decision theories.

Newcomb's Paradox

You, a mere mortal, encounter P, some super smart alien.  Or maybe it's a supercomputer, or maybe a god; versions of the paradox differ on this.  P comes up to you and says: "I have a deal for you.  I'm going to give you two boxes--box A, and box B.  Box B is transparent, and you can see $1,000 in it.  You can't see what's in box A.  I'm going to give you two choices.  The first is to take box A--you get whatever is in it.  The other choice is to take both boxes--you get box A, plus the $1,000 from box B."

So, you ask, why don't you take both boxes, getting the free $1,000?  Well, says P, there's a catch: "I have predicted whether you will take one box or two boxes."  (Or maybe I've simulated all of the atoms in the universe, or maybe studied your psychology, or maybe something else--versions of the paradox differ in how P knows how many boxes you're going to take.  But however he knows it, you believe him; maybe he has, in the past, predicted everyone who's taken this challenge successfully.)  "So I know what you're going to do", says P, "and before you arrived I decided how much money to put in box A.  If I predicted that you were going to take only box A, I put $1,000,000 in it.  Otherwise--if I predicted that you were going to take both boxes--I left box A empty."

"So", says P, "How many boxes do you want to take?"

Sunday, November 11, 2012

Elections and the Future

The Democrats' victory in the 2012 elections--primarily President Obama's reelection but also the Democratic caucus in the senate growing by three senators*--has caused a fair amount of hand wringing among conservative circles about the future of the Republican party--the new fashion in political circles seems to be guessing which of opposition to comprehensive immigration reform, opposition to gay rights, and opposition to tax hikes for the wealthy will have been felled by the 2012 election.  I agree in large part with the long term trend of American politics, but I think it's important to keep it in perspective.

Tuesday, November 6, 2012

Election 2012 liveblog

Welcome to the Measuring Shadows 2012 election liveblog!  We'll be live chatting on the widget to the right, and writing longer comments on this page.

Monday, November 5, 2012

Lame38, and liveblogging the election

I will be liveblogging the election tomorrow night on this blog, starting with some posts early in the day but picking up around 7pm.  But why should you follow this blog?

Because there is a crisis waiting to unfold tomorrow night: what if polls close have just closed in Virginia but Nate Silver hasn't updated fivethirtyeight yet?  How will you know if Obama's chances of winning the election have jumped to 92.5% or plummeted to 91.5%?

It is in the hopes that this catastrophe may be averted that I unveil Lame38: a shitty version of 538 that I promise to update frequently.  What is Lame38?  Well, I took all of the projections from fivethirtyeight, created a simplified model, and will run it with updated results--giving a real time projection of what Obama's chances of winning are, incorperating both states that have been called and the actual votes from noncompetitive states.

Q: Why is this better that 538?
A: Well maybe 538 will take like 15 minutes to update but Lame38 will only take 5 minutes.  That'd be pretty cool, right?

Q: So it's just a shittier version of 538?
A: That's kind of a rude question.

Q: What information does your model incorporate as the night goes on?
A: I incorporate not just what states have been called, but also popular vote results from each state to try to estimate biases in the projections.

Q: So how does Lame38 work?
A: I took the projected vote differences and errors for each state from fivethirtyeight, as well as the standard deviation of the national popular vote.  For each simulation I will sample a national popular vote, bias each state by its difference from the projected national popular vote (to estimate national bias), and then sample each state's vote.  I then run about 1,000,000 simulations.

Q: So, what are the odds Obama wins the election?
A: Well, 538 says 92.2*%, and Lame38 says 93.5%, so I'd say the answer is about 92.2%.

Q: What else will you be blogging about?
A: Senate races, the race for California's 15th congressional district, and whatever else is on my mind.

Q: Will this be a great liveblogging, or the greatest liveblogging?
A: I'm shooting for "worth reading".

I hope to see you there!

*When I ran the simulations; 538 has since updated to 92.0%.