The 10th Man: How I Disproved Efficient Market Theory by Being a DJ

Yes, DJing is a hobby of mine. I’ve been doing it for about seven years, picking it up late in life. Call it midlife crisis number one.

I have had a pretty fun DJ career, all things considered. I’ve played in a couple of really famous clubs, and I’ve done all kinds of private parties—parties where people have gone nuts. I post my mixes online for people to enjoy. I’ve spent way more money than I’ve earned (especially on music—I pay for all my tracks). But it’s been worth it.

Most people here are probably dimly aware of the electronic-music phenomenon that’s been happening worldwide, but they might not know or understand what DJing is all about. A DJ plays recorded music live. That’s it. He chooses the order of the songs. In the old days, this was done via vinyl records, and there was a great deal of skill involved in adjusting the speed of the records to match the beats. Nowadays, you can get a computer to do that for you if you want, though the annoying purists will hate you for it.

None of this sounds hard, right? Why do people consider it an art form? Well, first of all, the DJ is a curator of music. This is more difficult than it sounds. It’s not like pop music, where the same ten songs play on the radio on a loop. If you’re into underground music like me, you’re digging deep to find great tracks that nobody has ever heard before. In the old days, this would entail spending hours at the record store. Now you can do it online. But it’s still a lot of work.

Is the curator an artist? Yes and no. Think of the museum curator—he decides what paintings to hang and what order to hang them in. A DJ does the same thing. It’s an art form that is not fully appreciated. Back when I lived in New York and went to clubs, there was a progressive house DJ named Zack Roth who used to warm up for all the big trance acts that came to town. I was obsessed with being a warm-up DJ. The warm-up DJ goes on first, around 10 p.m., and plays dark, deep stuff, gradually bringing up the energy (and the tempo) until he’s whipped the crowd into a frenzy just as the main event comes on. Zack Roth was the master.

All this is done by playing songs in a particular order. Amazing!

Why does this work? Well, it works because you have memory. You may not notice the transition from one song to the next, but you will notice the increase in tempo, the rise in energy. Over the course of a few hours, you will have noticed that you started at 1 and ended up at 10.

If you had no memory—if you had the memory of a goldfish and you couldn’t even remember the last song played—DJing wouldn’t work. You could play the songs in whatever order; it wouldn’t matter. The only song that would matter would be the one you were currently playing.

Path Does Matter

There is a concept in finance known as path dependency. Like, the price of a security goes from A to B over time. Does it matter what path it took to get there?

In the derivatives markets, this question is of crucial importance. There are some exotic options, like lookback options, whose price actually depends on the path the asset takes. But with plain-vanilla puts and calls, path doesn’t matter—all that matters in the pricing of the option is where the asset is currently.

It seems obvious that path would be super important, though. Think of it this way: right now the S&P 500 is at about 2,000. You know what path it took to get there. It went down to 666 in the financial crisis and has basically gone up for six years straight, until today. People are pretty bullish, right? The market has been going up every single year, for six years.

But what if the market had gone down to 666, then up to 4,000, then crashed to 2,000? This is a stupid example, but not really. Would people feel the same with the SPX at 2,000 if it had taken this path, crashing 50%, rather than the previous one, where the market went straight up?

Of course not. If the market had arrived at 2,000 by going down 50%, nobody would be bullish at all.

Path really matters. Here’s another example.

The Gas Tax and Path Dependency

Gas prices were really high over the last several years. Even if you’re not a financial expert, you can probably tell me the approximate path of gasoline prices over the last 20, 30 years. Because you remember.

So gas prices were high for years, running up to $4 a gallon and more—until they suddenly crashed, dropping by over 50%.

Did you see what happened next?

Politicians started calling for an increase in gas taxes. Why? Because people were used to high gas prices and could easily absorb an extra 10 cents on the gallon.

What’s interesting is, if gasoline had been $2 all along—or, say, it had been at $1 for 10 years and then had gone up to $2—there is no way politicians would have been calling for gas tax hikes. People would be furious. So gas prices are path dependent!

Well… so is everything else.

EMH Is Dead

The Efficient Market Hypothesis is the idea that all information (including the path of previous stock prices) is reflected in current stock prices.

Even in the age of the Internet, this is not true.

There was a 60 Minutes episode on curing cancer last Sunday. The biotech guys have known about this for years. The information is publicly available, but unsurprisingly, most people don’t make the effort to learn about that stuff. It takes time for information to travel, sometimes a long time.

Most options are priced with similar assumptions. They’re modeled on something called geometric Brownian motion, which describes the behavior of a particle suspended in gas. The particle has no memory. It is path independent. But unlike the particle, the market has memory. People have memory. And as we demonstrated, it is path dependent. So its behavior will not be a true random walk.

People are slowly learning what the quant guys have known for years: the market is not random, and you can profit from that. There would be no commodity trading advisors (CTAs) if the market were random. But the guys who have disproved market efficiency are making too much money to bother filling out a Nobel Prize application.

This is the problem technical analysis claims to solve. The technical guys are half right. The market is path dependent, yes, and they have constructed a set of rules to describe this behavior. Problem is, sometimes the rules work… sometimes they don’t. Technical analysis is most helpful as a guide rather than gospel.

This is a roundabout way of saying that the efficient-market people who tell you that you can’t beat the market so you shouldn’t try—well, those people suck. Don’t hang around with those people. Deep down, I have always felt that you should try. Anything worth doing is worth doing well.

Jared Dillian