From the Chicago Tribune: Father arrested for forcing son to wear Green Bay Packers jersey

Upset that his 7-year-old son wouldn't wear a Green Bay Packers jersey during the team's playoff victory Saturday, a man restrained the boy for an hour with tape and taped the jersey onto him.

Awesome.

If you don't know who Ezra Levant is, check here. If you don't know what he's talking about, check here. (And then get your head out of the sand.)

"Whatever offends you...I reserve the right to publish it for whatever offensive reason I want."

To quote my favorite online news commentary site: THIS.

Christopher Hitchens, "The Case Against Hillary Clinton."

Revamped, revitalized...

~clink to Savage. He's rekindled the (fake) additction to gambling. Thus far:

• I lost my ass on OSU;
• Won huge cash on NE v. Jacksonville in that I both: bought shares in the correct spread and short-sold the Jags;
• Lost MY ASS on Indy;
• and just got a massive payout on Mitt in MI.

Fair warning, this post includes math.

Ok, now for the two of you who have stuck around... the reason I bring this up is because of the ads I've seen for a new TV show. Now, normally I barely see commercials since I bought TiVo (aside from the personal computer, my favorite gadget in the world). But moving into a new home means that TiVo's not been hooked up, and if I want a respite from unpacking, or to see the games (fucking OSU -- seriously, people, come on), I'm stuck with the ads.

The newest game show to hit prime time is something called "The Moment of Truth". If you don't click through to read about it, and I don't blame anyone who doesn't, the idea is this: sit someone in a chair, strap them to a lie detector (polygraph) and ask embarrassing questions. You win money if you answer "honestly". Why do I put that in scare quotes? Well, that brings me to the math. Almost.

Polygraph machines have been around a long time, and are a key feature in things like police investigations and security clearance investigations. Which means they feature prominently in the prosecution of justice safeguarding of secrets. But do they work?

Not the way they'll tell you they work. Here's the problem: detecting a lie isn't easy, and the test relies on subtle physical changes that are believed to be cued when you lie. That they may be cued when you are stressed (in times when, for instance, you may be being questioned about a crime or are trying to get a clearance to get a job) is an open question, but we'll skip over it for the sake of simplicity. When asked, polygraph examiners will cite how "accurate" the test is as a sign of how well it works. This is bullshit. It may be great interrogation, but it's mathematically worthless. The problem is that a test like this has not only a chance of being right, but also a chance of being wrong.

Bergeron will likely know about an analogous situation: You go to the doctor and have a test done to see if you have Terrible Horrible Impotence Nurturing Genetic Enzymes. The test for THINGE says it is 95% accurate. So, your THINGE test comes back positive. What then, is the probability that you have THINGE?

Did you say 95%? Well, I'm guessing most people go to the point of "Well, I would, but this seems like a trick question." Give yourself points if you did. Because it is NOT necessarily 95%. We need to know more things. That 95% rate is essentially the true-positive rate. That is, 95% of the time it will give a positive reading in the presence of someone who has THINGE. But it will also have a false-positive rate. In this case, make it 5%. And we have to know the underlying rate of the disease in the population. Let's make it easy and pick .10; that is, 1 in 10 people in the entire population have THINGE. So, you got a positive test back--what is the probability you have THINGE?

It's this: The chance that the test comes back positive GIVEN that you have THINGE times the chance that you have THINGE, all divided by that plus the opposite (a positive test result given you don't have THINGE times the chance that you don't have THINGE.

Let's do the best part of math and shorten all that:

• P(p|T) is the probability of a positive test result given you have THINGE, or, the "test accuracy"


• P(T) is the probability you have THINGE, or the .10 rate


• P(C) is the probability you are clear of THINGE, or 1-P(T) = .9


• P(p|C) is the probability of a positive test given you DON'T have THINGE, or, the "false positive" of .05

Ok, then: P(p|T) x P(T) is .95 x .10 = .095
And: P(p|T) x P(T) + P(p|C) x P(C) = .095 + (.05 x .9) = .14.
Finally: .095/.14 = .6786.(*)

With a positive test result, on a test that is 95% accurate, you have a 68% chance of having THINGE. Not good odds for you, but much better than 95%, eh? Which is the reason doctors have multiple tests and look for multiple possible indicators for a condition.

Back to the main point: the exact same issue holds for polygraphs. Proponents claim they are "close to 100% accurate", while critics say they are only around 80% accurate. This is a huge gap, though it might not look like it.

Let's say you're being screened for a security clearance. They want to know if you've murdered anyone. How confident can they be that they weed out a murderer? Well, just go through the calculation again. This time, we'll use the US average rate for homicide, and we'll use the 2005 18-24 grouping, which is the highest rate: 26.5 out of every 100,000, or .0265% . And, lets say the test is really good: 95% accurate, with a 5% false positive rate. Finally, needless to say, you answer no. (We won't say whether or not you did yet.)

Doing the calculation as above, we get that the probability that you murdered someone given that the test is positive (show's you are lying by answering "no" to the question of whether you have ever committed homicide), is around 34%. Barely better than one in three, assuming that nothing other than you lying affects the results of the test.

In the case of the TV show, things are even worse. How do you get an underlying distribution of people who are racist, who have thought about their boss sexually, or how do you judge if someone seriously considered divorce or if they just had a shitty day?

In the end, this is just idiot people doing anything for money. And it's their choice, so good for them. The real disservice is that even the people behind the show, even the people who administer the test, may have absolutely no idea how bad their test actually is.

(*) For the interested, this is simply Bayes' Theorem.

[Excuse the muck if you saw this post before -- and if you're on RSS. I just burned a couple of brain cells trying to get decent info up here and it was shite. This should be better.]

I cannot, for the life of me, understand how Inklings does its login. If you sign up by going through our markets, you will need to sign up for each one. AND you'll have to sign up to trade on the other public markets. I've made tons of fake accounts now, and I can't sign up through either of the two markets we have and then get the login to work for other public markets. Same with going in the other direction.

If you're interested in trading even fake money in other markets, use InTrade since they have a decent play money section.

MORE UPDATES: I'm a big baby. The site is decent, and does a service by enabling people to make their own markets fast. But it still frustrated the crap outta me.

So, it turns out Adam Siegel, the founder of Inkling, is, like, a decent guy, and didn't rise to the bait of my snark and provided some useful info via email. I'll pass on the bulk of his suggestion:

There's no need to open up new marketplaces for each market. Consolidating it all under one page (i.e. inklingmarket.mint400.net or something) is probably the best way to go about things so that users can access them through one ID. Second, the OpenID tool is available for people if they want to use one ID across multiple sites.

A couple of notes, from my email back (I don't pretend to think the guy has enough time to keep checking the Mint):

1) Ok, I get that having one site makes sense. But as a group blog it's not all that convenient to coordinate all ideas for posts through one person. That this ended up on Inklings was a whim of mine; this shouldn't tie another contributor to the blog to using my ID and login, or even to create new parts of the site. But this, of course, is the push-pull of one site's ease versus another's (it's easier for Inklings if we just consolidate).

2) One of the big issues with prediction markets in general is that the insights they provide are degraded if the market is too "thin", that is, if not enough people trade. Which means we want as many people trading as we can get. I opened these up to the public because the few people I could allow access to through email would be paltry compared to what is really needed. Major roadblocks in moving from market to market means a lot of people ditching from frustration, thus thinning out markets for everyone.

3) After a couple of years in the website biz, I can claim some insight on user issues. The lack of a good standard for third-party sign in systems (e.g. TypeKey, OpenID...) still means that I have to maintain multiple IDs and passwords (or, in a move that makes the security freak in me twitch, use the same ID and password for everything). Sure, you have to sign up for an account through Inkling, but I don't think a lot of people come to it with an OpenID. So they can either accept the usability of Inkling, or go sign up for yet another account, perhaps just so they can trade across multiple markets. This splits the users attention and adds a weird layer of interaction that some people might find off-putting.

Anyway, like I said, Adam provided feedback and insight to some random guy on a blog. We here at the Mint have a pretty strong policy of not only pointing out when something pisses us off, but also recognizing real stand-up acts. Thanks, Adam.

Savage made the first entry in this vein below. After hearing Huckabee's victory speech, I want to know how many think Huckabee will be the nominee from the Blue camp?

This market, as well as those we create in the future, will be posted here: Mint400 Inklings.

Ride On.

In the post below I made a bet: Barack Obama will be on the Democratic Party ticket for 2008, either in the slot for President or Vice-President. Seed took me up on that (and a slightly expanded) bet. Which is fine. But this happens to fit another one of my personal interests: prediction markets.

Prediction markets are trading mechanisms for options on future events, times, values, etc. In this case, the future being traded is Obama being on the Dem ticket. You can buy shares of either Yes or No. The idea is that you'd buy the shares in the quantities that correspond to your conviction about the event. It's just like actual trading (you can buy shares of BOTH, if you want, to cover), with the added benefit of being able to read the price of the option as a percentage. That is, if the shares were selling at $50, you could read that as saying the traders in the market believe, on average, that there is a 50/50 chance on the outcome.

I've built a market like this at the site Inklings Markets, which lets you set these things up in a snap. And it's open to the public:

Obama on the Dem Ticket Market

I've not used this much, but I think if you head over and register, you can start trading. You get $5K in play money. Spend it all on the Obama market, or elsewhere. I'd love to see a ton of people involved, so spread the word if you're interested.

Jets to resume beer sales in '08

The Jets instituted Sunday's prohibition after increased security at their prior home game failed to dissuade hundreds of men from imploring women at Gate D to expose their breasts at halftime.

Mercy.

I happen to really like issues associated with voting. The mechanisms and procedures, as well as how or why people vote.

Still awake? Good.

Since it's that time of year, I thought I'd put up something about the caucuses that are going on in Iowa tomorrow. First, note that the Iowa caucuses have a decent history of picking the people who will ultimately be the representatives for the two parties. It's not a lock, though, and in years where the field was relatively uncertain, Iowa doesn't always do as well. But look to the top people out of Iowa to be pretty much be the two to four people we'll be hearing about the most for the next year.

The next point is how the caucus is actually conducted. For the Repubs, it's pretty simple: you show up, listen to some electioneering from friend and neighbors, and then put your choice down on a piece of paper and dump it into box. For the Dems, however, things are a bit more convoluted. Indeed, I'd lump this in with a "too clever by half" sort of designation.

Caucusing for the Democrats functions, essentially, like a Borda Count. But not exactly. So, say you're a proud D in Ames, Iowa and looking to caucus. You get out of work and head to your local station with some thoughts in mind about your choice of candidate. Appealingly, all you'll be required to do is stand in some part of some room somewhere to indicate your choice. Like Clinton? Over there by the chalkboard. Prefer Obama? Stand by in the back left corner. Whatever. Then, while you're standing there, you get treated to the best rendition of a used car salesman your friends can drum up. They plead, beg, wheedle, cajole, criticize, and harangue you in order to get you to move to their group and support THEIR choice. Thirty minutes of that and time is called, and the numbers are counted.

Which is where the Borda count analogy starts to show up. If one candidate doesn't get 15% of the total vote, they are eliminated for "non-viability". When those non-viable candidates are weeded out, everyone goes through the choice process again. Now, however, there are fewer candidates for the same amount of people. In a Borda count, you would write down your preferences by indicating a vote for each option with (generally) non-repeating integers where your top choice gets the largest value. With five people A-E, it might be:

B: 4
C: 3
A: 2
E: 1
D: 0

and something similar for everyone else. In totaling up scores, you simply add the values for all the candidates to see who wins. (Ties are possible in the basic framework, but this can be avoided with some effort.) Consider some of the problems, too. Say that 10 people vote, and of those 5 vote for B with 4 points, and the other 5 vote for B with 1 points, while everyone voted for C with 3 points. B, preferred as top choice by more people than C, loses out in terms of points (25 to 30 points). B could even be counted as winner; the point being that the count process can result in some outcomes that people might not appreciate.

More interestingly, Borda counts are, like most other voting schemes, subject to manipulation and strategy. Perhaps you really want B to win, but know B would lose to A when compared against only each other. What you want, then, is to help make that comparison not happen. Well, why not choose someone B would do well against, and make that person your insincere top choice? Put C on top, B second, and A all the way down at the bottom, and you do your level best to keep support from A, while making it possible that B wins entirely because B can beat C in comparison. Think of this in connection with Iowa. So, you like Clinton, but think she'll never beat Obama in a one-on-one fight. Well, when you go caucus, why not stand with the Edwards supporters? After all, he's an airhead lightweight that would get clobbered by Clinton's political machine. But if Edwards gets second or first in Iowa, it may make Obama go running home to cry until he's second on the ticket [Editorial Aside:] I'll take on all betters willing to say that the 2008 Democratic Ticket DOESN'T have Obama in one of the two slots. Seriously. Before we even get through Iowa, I'm willing to say that the man has assured himself a seat in the White House in 2009. Not only is he such a driving force in the Dem party that I think he's got a shot at top billing, there's just no way that Clinton, if she does make it, can choose ANYONE else. The headline alone, A Woman For Prez and the First Black Man in the WH, would be enough to guarantee Obama a spot. But add to this the list of halfwits and knuckleheads the Repubs are putting up this year, and I really don't see it being an R in 2009. End Aside. Or, maybe you're looking to really bring someone else down a peg. Or you just want to muddy the waters a bit, and keep someone in the race you wouldn't care much about if it weren't ultimately going to help someone you do like. So you go stand in the corner for Richardson, hoping to keep him viable and his campaign alive.

So while you're suffering through all the talk about what's going on tomorrow, just wonder how many people are actually voting with the intent of pushing candidates this way or that. And from this the winners ride political "momentum" (which I think is just a measure of how much the media favors one person over another and thus how much coverage they get) to other states and possibly to the nomination. Of course, this assumes people can think through the implications or do some of the math on their own.