Nate Silver Defends His Model

Mr. Silver has been the focus of a lot of attention lately, but mostly it has come from people who don't understand his model or probability theory. It's not outrageous that his model should show President Obama with an increasing probability of being re-elected: under a Bayesian probability theory, probability should continue to approach 1 until it reaches it. If nothing disturbs existing trends in the polls, the probability should be well above 90% by Tuesday morning.

This is because his model is a whole world to itself, and within that world the number of things that can change the probable outcome are declining every day. It is exactly like the program that calculates the odds of a Texas Hold 'Em hand winning or losing when you watch the World Series of Poker. Every time a card is turned over, if it doesn't materially affect the odds in favor of the challenger, the odds of the high-hand holder go up. This is because there are fewer cards left that might change the game for the low-hand holder.

Thus, Mr. Silver is to a large degree correct. If the state polls are correctly modeled, and if no new data out of line with the existing data is introduced, time is grinding away the opportunities for anything different to occur.

Nevertheless he makes a huge error.

But many of the pollsters are likely to make similar assumptions about how to measure the voter universe accurately. This introduces the possibility that most of the pollsters could err on one or another side — whether in Mr. Obama’s direction, or Mr. Romney’s. In a statistical sense, we would call this bias: that the polls are not taking an accurate sample of the voter population. If there is such a bias, furthermore, it is likely to be correlated across different states, especially if they are demographically similar. If either of the candidates beats his polls in Wisconsin, he is also likely to do so in Minnesota....

My argument, rather, is this: we’ve about reached the point where if Mr. Romney wins, it can only be because the polls have been biased against him. Almost all of the chance that Mr. Romney has in the FiveThirtyEight forecast, about 16 percent to win the Electoral College, reflects this possibility.

That "sixteen percent" chance that the assumptions are flawed is entirely bogus. The model can account for sampling error of the "plus or minus three percent" variety; there's no problem with that because you can give it a percentage using known methods. But there's no way to know what the odds are that a flawed assumption is making the data itself unreliable.

That's the one thing that the model can't actually measure. Any attempt to estimate it is a completely unscientific guess. Investment bankers and pollsters are each a class: they make slightly different guesses, based on their position and what they want to achieve, but they inform themselves based on talking to and watching each other. If they're wrong, it's as likely as not that they're almost all wrong.

There's simply no way of knowing what that means for the model, because the whole model is built around data shaped by their assumptions. It's like trying to guess what the odds are that Elvis is still alive in a nearby possible world: there's no way to put a number on it, because the world in which Elvis is alive is a world built out of entirely different facts.

If the pollsters are biased, the facts of the true world and the facts of Mr. Silver's world are simply not the same at all. His model won't just be wrong in some way that can be estimated and worked into the model. It'll be so wrong that it simply can't be applied to the actual world. Any resemblance between his world and ours will be accidental.

That's How We're Gonna Beat 'Em, Butch:

A German Neo-Nazi went to Afghanistan to train and fight with the Taliban. At last, he had to give it up and come home, saying he made a big mistake. Why? According to him, his wife made him do it. She was very upset by the lack of creature comforts, the bad treatment of women, and the fact that they couldn't hold hands in public.

Oh, and also:

His catalogue of complaints over his poorly thought-out decision included his fellow combatants' drug habits, a lack of hygiene, contracting hepititis A and his friends being horribly killed by Pakistani forces helicopters.

Well, yeah, that too.

Bacteria are your friend

From Rocket Science, an article about a "self-healing"concrete that uses limestone-secreting bacteria activated by water that seeps into tiny cracks.

A Constitutional Right to Hunt

Georgia did a version of this some years ago, but apparently Kentucky has come up with a twist: ordering regulatory agencies to consider hunting the preferred means of animal control.

The thing about a "personal right to hunt" is that it is one of those rights that require provision from us to you. If you have a personal right to own a firearm, if you want to go and own one you can, but you have to buy it. The state is merely forbidden from making it impossible to exercise that right.

If you have a personal right to hunt, however, that means there must be some way of exercising that right. In Georgia, for example, we have many public lands that are owned by the state and set aside for the purpose of public hunting. I imagine Kentucky has similar lands.

It's not a big deal, since the opportunity costs of such lands are often low, and as they are also useful for many other good public purposes (such as hiking and camping) outside the hunting season. The hunting area in the Dawson Forest, bordered by the Amicalola river, is a beautiful place. I've never gone hunting there, but I've hiked the river many times.

Still, it's an interesting point because it's one of the few so-called "positive rights" that conservatives generally support.

I Have Voted

Went over to the county seat today, since I was out that way on other business, and took part in early voting. The line was surprisingly long, but even more surprising was its demographic composition. This county is nearly ninety percent white and about eight-and-a-half percent black, but fully half of the folks who showed up to vote while I was in line were black.

My guess is that any slippage in Democratic voters' enthusiasm won't be coming from the black community this year. Now, of course I don't know how these ladies and gentlemen were voting, and am merely assuming based on the historic record that most of them were probably Obama voters. Still, assuming that record holds, the President can count on a surprisingly strong showing in this part of rural Georgia.

Models and their perils

A good article on RedState on polling methodology, with comparisons to modeling of climate, housing markets, and baseball:

Consider an argument Michael Lewis makes in his book The Big Short:  nearly everybody involved in the mortgage-backed securities market (buy-side, sell-side, ratings agencies, regulators) bought into mathematical models valuing MBS as low-risk based on models whose historical data didn’t go back far enough to capture a collapse in housing prices.  And it was precisely such a collapse that destroyed all the assumptions on which the models rested.  But the people who saw the collapse coming weren’t people who built better models; they were people who questioned the assumptions in the existing models and figured out how dependent they were on those unquestioned assumptions. . . .

I was mistaken in one of my comments below about the "Unskewed Poll" methodology. They do attempt to conform poll responses to a turnout model; they just use an unusual model.

Aid for the Northeast

The original headline on this piece from Staten Island was "We're going to die."

Clearly the government's efforts here are inadequate and misplaced. Given the gasoline shortage in the wake of the storm, a lot of the private options for help are limited because we can't all truck up there with food and supplies. However, if you are able and wish to help, you can donate to relief organizations already on the ground.

Catholic Charities and the Salvation Army are two that are reliable. Both take their religious obligations to help the poor and afflicted seriously, and do so without regard to the religious status of the people they are helping.

If you know of other options that you can recommend, please mention them in the comments.

Negotiations

Last week I got a letter from the bank saying that I owed them $73 for rental on a safety deposit box that my wife apparently took out while I was in Iraq. It was free at the time, but apparently there had been some unannounced change in the policy. I called them to ask that the unannounced fees be waived, and they were intransigent on the phone. "Well, I'll come in and discuss it with you when I bring in the keys," I said.

At the end of that discussion-in-person, they agreed that they would waive the whole $73 and furthermore pay me $4.60 in return for accepting my keys and canceling the contract. I'm not sure if it was my charm, courtesy, or motorcycle jacket.

Unskewed map

My husband suggests that, if this electoral map from Unskewed Polls is even close, David Axelrod will be obligated to get a full body wax.  On live TV.

Of course, I have no idea which polls are accurate.  I'm hoping that most of them are over-weighting the Democratic vote by mistakenly applying data from the 2008 rather than the 2010 races, an error that the "Unskewed" site claims to be avoiding.  We'll see soon enough.

Questions no one in the media is asking

No, not about Benghazi.  I've about given up hope on that one.  What I want to know is whether Senator Menendez (D-N.J., appointed to fill Corzine's vacated seat) is underpaying his male prostitutes as callously as he is the female ones?

Happy Halloween

Disaster prep

One of the first things the emergency response corps does here in implementing a mandatory evacuation is cut off the power, water, and gas to the most exposed shoreline areas.  I'm surprised to see reports of devastated communities (such as Seaside, NJ, in the video below) burning down and gas lines still clearly attached.  On the TV news just now, people were discussing difficulties in finding personnel to shut off the lines.

The new craftsmen

I approve the trend in pumpkin carving.

Memento Mori

Dr. Mead is moved by the storm to musings of the kind that occupy my mind so very often. He has written a good piece, just the sort of reflections that these encounters ought -- once we have done all we can do to prepare or to help -- to provoke in us.

The inimitable Ramirez

Very bad on the coast

Massive fires in Queens:




And -- is that thing really a shark?



This substation explosion really would have gotten my attention.  I can't believe people are walking around in that water.  There's way too many buried electrical lines in New York for that kind of thing.

New York Submerged

Zero Hedge has many pictures, including this shot of Ground Zero.  The tunnels are filling up.

On the Night of the Hurricane

Nothing tonight, save a prayer for those under the storm. First among those for each of us is the one most beloved who lies under that storm; but for all of us who are Americans, among the first must be these.

On the Polls

One of the more interesting stories of this election has been the strangeness of the polls. I don't simply mean the way they are being interpreted by Democrats as clear evidence of a definite Obama victory, while at the same time by Republicans as suggesting a huge Romney landslide. That's to be expected: even if you're not spinning polls at all, it is natural for the partisan mind to weight explanations and interpretations that give favor to their side.

No, what I mean is better brought out by this Washington Post piece. It's an argument against interest on the House elections, but think about what it says.

President Obama remains at least an even bet to win reelection. Democrats are favored to hold on to the Senate — an outcome few prognosticators envisioned at the beginning of the year. And yet, with a little more than a week to go, the party holds almost no chance of winning back the House.

“They called the fight. It’s over. We’re going to have a House next year that’s going to look an awful lot like the last House,” [said] Stuart Rothenberg[.]

In other words, during the worst economy since the Great Depression, in spite of the deep unpopularity of every political branch, in the face of a government so badly run that their idea of smart budgeting is to dive off the fiscal cliff rather than pass a budget... the polls suggest that Americans will use the election to endorse almost exactly the same government for the next two years.

There are two possibilities here. One is that the polls are fundamentally wrong: some aspect of their methodology is distorting the picture badly. The other possibility is that we have structured the political system in a way that is too stable for its own good.

If it were only the Presidential race, it could be something about the candidates. But it's not: the polls suggest stability across the board. It could be that some combination of gerrymandering, ideology, and the like has brought us to the point that most Americans no longer face a real choice at the ballot box. There's a candidate they have to support as the lesser of two evils, because the other guy is somehow deeply against the things they care about. If that's the case, then even in the face of a government as badly run as this one, the democratic mechanisms can no longer make a significant adjustment.

Is that the case? Well, we had wave elections in 2006, 2008, and 2010. It's hard to believe that the picture has since solidified in that way.

Go, Mighty Bulldogs

Now, some of you might have seen the Georgia v. Florida game today, also known as "the Cocktail Bowl." Florida was #2, and a victory would ensure them a place in the SEC Championship in Atlanta. Georgia was the under-Dawg.

Most likely you know how it went.



That is all.