Alex Whines

Yesterday, I asked my brother whether he planned on watching the Super Bowl.  He hadn’t made up his mind, but suggested that having silly things to bet on would make the game more interesting. That got me thinking about silly bets that could make the game more interesting.

Since I promised a hot investment tip, I won’t bury the lead any further: if you watch the Big Game this sunday, you should bet some chump that the total score of the game will be an odd number.

When I first started to consider this, my inclination was that the score should be even around half the time or maybe a tiny bit higher.  Since almost all scores in football are either 3 or 7 points, games should go back in forth between odd and even totals with every score change.  Since I couldn’t think of a good reason that there should be more games in which teams score an odd number of times than games in which teams score an even number of times, it should come out to pretty much 50% chance of an odd score.  (the tiny bit higher comes from the fact that some scoring plays are worth an even number of points.  If there were only even scoring plays, the total score would always be even, and this fact (possibly) tips the chances of a even-pointed game above 50%).

That is what you should explain to your mark, to get them to practically give away their money.  I would recommend keeping the stakes kind of low so as not to tip them off to your ruse (also: avoid twirling your mustache villainously*).

But you’ll know better.  There are 256 regular seasons NFL games each year.  Thus, 1536 games have been played over the last six years.  In those 1536 games, the total score has been odd in 857.  That’s an astounding 56% of the time.  Furthermore, in all six seasons, there have been more odd-pointed games than even-pointed games.  Suggesting this is no fluke.

That means, your expected gain for each dollar you wager is 12 cents.  That’s huge, dwarfing any expected return you’ll get from a wager through a sports book, bonds, or mutual funds.  Really, this is a can’t miss opportunity.

Although I believe the statistics, I’m not entirely certain on the mechanism.  It might have to do with when to do for two, or the fact that teams try to break ties with field goals, but I don’t have any excellent insight there.  But I don’t understand how audio can be encoded as a digital signal, and my ipod works just fine, so you probably want to take my advice.

*Unless, of course, you have been working a longer con, in which your mustache twirling has been consciously established as a tell, when in fact it is a reverse-tell.  In such situations you should offer the opposite bet that the game will end with an even number of points scored, and twirl your mustache villainously while you do.  The mark, believing this to be your tell, will insist on changing the bet around.  You will pretend to reluctantly agree, and your reverse-tell mustache twirl long con will be a success.

This econ paper that Freakonomics linked to is pretty darn cool.  It’s called “To Review or Not to Review? Limited Strategic Thinking at the Movie Box Office” (authors: Alexander Brown, Colin Camerer, and Dan Lovallo).  It suggests some interesting results:

1. Movie critics actually matter. The paper compares wide-released movies that were screened for critics before opening with “cold-opened” movies, aka movies that studios know are so crappy that it’s better not to have them reviewed at all.  Controlling for all the things you could thing of (metacritic score, budget, genre, rating, and a gaggle of others), they find that not letting the critics see a movie actually helps the box office gross.  This means that bad reviews matter.  Which I find quite comforting.
2. People are stupid. If all moviegoers were rational, in a repeated game like this, opening cold should never be the right decision.  It should signal: “This is a movie that is going to the get the worst reviews possible.”  This suggests that isn’t what happens**.  Rather, having not read anything bad about the movie, the movie going public knows it must not be great, but maybe it’s pretty good.
3. Some people aren’t as stupid; they work of movie studios. We read stories all the time about how studios mishandle the promotion of certain movies, and those criticisms always seem pretty valid to me, but this paper suggests that studios have a decent grasp on movie goers (and critics), or have some sense when to cold open a film.  The paper also mentions that the practice of cold-opening has increased in the recent years (although I’m curious if more bad movies are being made each year).  Also, the model they are using only requires some people to be stupid, not every single person that watches movies.
4. Critics aren’t that important. One of the most interesting results is relegated to a footnote.  The “cold open premium” persists after the opening weekend.  This doesn’t make sense at first because now reviews exist (and they are shitty).  The theory the authors put forward is that moviegoers use information on first-weekend gross to decide whether to see the movie the second week, and that this must overshadow the information on critical reception that they also have.  This would be perfectly rational if the first weekend watchers were rational, which we’ve established is not true.

The paper is very cool, even if I didn’t need this much data to know that people who went to Legion are idiots.  And perhaps the real takeaway is that the stupid moviegoing masses are slightly smarter than I give them credit for.  We’re dealing with a fair bit of informed decision making here (they pay attention to reviews and follow box office grosses when deciding what to see), but not quite enough; as the title says: “limited strategic thinking.”

*Here’s why this happens: Suppose movies are all scored on how much critics like them from 0-100 (and let’s say the scores are evenly distributed).  Suppose Wilma is releasing a new movie, “The Dancing Zombie”, and Waldo is thinking about seeing this movie.  If there are no reviews, Waldo should assume (based on the distribution of movie scores) that “The Dancing Zombie” is a solid 50.  But then Waldo thinks: “Wilma knows I’m going to think it’s a fifty, so if she thinks it’s higher that 50, she would have let critics review it.”  So then, Waldo must assume the movie is somewhere between 0 and 50.  25 is a good guess.  But then Waldo thinks: Wilma knows that I’m going to think that she knows that I’m going to think it’s a fifty, and hence she it must really be less than 25.”  If both parties are rational, and both parties think the other is rational, a limit approaches 0, so any review is better than none.

**But in reality, the prospect of seeing a zombie dance clouds Waldo’s rational perspective, and he goes and sees it anyways.  He probably sees the sequel too.

I’ve decided I’m going to find a word that ends with “-LESSNESS” while playing online Boggle.

There are 22 that have 11 letters in them:

‘agelessness’, ‘aimlessness’, ‘airlessness’, ‘artlessness’, ‘awelessness’, ‘endlessness’, ‘godlessness’, ‘gutlessness’, ‘haplessness’, ‘hatlessness’, ‘joblessness’, ‘joylessness’, ‘lawlessness’, ‘leglessness’, ‘raylessness’, ‘saplessness’, ‘sexlessness’, ‘sinlessness’, ‘sunlessness’, ‘toplessness’, ‘uselessness’, ‘witlessness’

There are another 61 at 12 letters, but I think those 22 are a good place to start.  Seems totally doable.

About a year ago, I made a few lines of program that simulates a boggle board to test how often a word should appear.  The simulation uses the letter distribution you’d get with the version of Master Boggle that I own (there were at least two (very slightly) different sets of dice that came with 5x5 Boggle).  By that count, “lessness” appears only once every 160 boards.

Then I beefed up the program, to check to see if any of the 22 show up on a given board.  According to my calculations (which might even be correct), I can expect to see an actual “lessness” word about once every 342 games.  Which isn’t exactly often.

But the thing is, I don’t have a lot of friends, and none of my friends want to play 342 games of boggle with me—because they will lose 342 times.  So I never actually get to use my physical board, and the online games use different letter distributions, and distributions I’m not entirely certain about.  My sense is that the “lessness” words will show up a little more often online, but still with some infrequency.  My guess would be that getting any 11-letter word occurs less than once each 100 boards, which makes the once every 342 seem actually pretty high.  Anyways, I’ll let you know when I find something.

natface:

end transmission

Well I got at least one right then.

The Mercator Project is not going to be the name of the secret cabal.  The Mercator Project is a visible, nonprofit funded by all the big map companies.  The stated purpose of Merc Pro is going to be to promote map-awareness internationally.  Outwardly, this takes the form of bringing maps and education-y programs to third world countries, especially those involved in some sort of history changing conflict.  Thus, those involved in MercPro travel unsuspiciously to all the places one would need to if they were going to alter the course of history so more maps were needed.

All the CEOs of all the map companies will serve as the board members.  This way, they can meet (to discuss which separatist movements to sell missiles to) without alarming anyone in the Department of Justice’s antitrust division. And this way, the name could have been thought up by someone being cutesy in some marketing department, as opposed to evil, secret world-runners. And maybe the Mapmaker even works at MercPro, but in the above-board, giving maps to poor people department.  Maybe not though.  I might have to think that one through a little further.

Also probably relevant

and I’ve reached a perhaps intractable problem concerning The Mercator Project, the Indiana Jones as a mapmaker action movie Ben and I aren’t actually going to make.

Let’s stay you’re in the map making and selling business, and you’re putting together a secret cabal to affect geopolitics (and hence the names of things on maps).  And assume that this is a big enough enterprise involving the exchange of enough money that you need to set up an legal entity to manage all of this.  I’m not saying this is publicly traded corporation (far from it), but you need a name for all the contracts and bank accounts and what not.

Now, we’ve already assumed you are clever enough to secretly orchestrate the course of history.  And since the whole point of setting up the legal entity is to separate this shady business from the legitimate map-related operations, I don’t think there’s any way you go and name the legal entity after something associated strongly and exclusively with cartography.  It’s just sloppy.

Just imagine, some CIA agent in the 50’s is going to find money coming from The Mercator Project to the Mau Mau and go, “Mercator Project.  That sounds like it’s related to globes.  Someone should look into that.”

And even if the CIA agent doesn’t connect the dots, I just think it would be unrealistic for the secret cabal to make the dots so connectible in the first place.  It would be a bush league mistake, and these guys are fucking pros.

—————

Completely unrelated to the above, I learned that the earth is not actually spherical.  Nutso.