How Do People Play The Trading Game?

“That’s cool,” a coworker said when I showed him the game. “What’s the point?”

“I’m not sure. We might not know until people play it.”

And we might not still! Digg called it “A simple game that will help you understand the stock market,” and it does implicitly purport to do something like that, but it is contrived and nonrepresentative of an actual stock market in a thousand ways. It assumes zero transaction costs, no leverage, and perfect liquidity (so no trade of any size is significant enough to move the price of the stock). It offers no real risk to compel loss aversion and lays only a thin veil across history to shield it from distortionary hindsight prescience.

Every player starts with $500, which is fully invested in the given stock with every tap. If we look at a sample of 13,000 people, playing 30-second levels, they ended up with $6.7 trillion (including winnings from their previous play-throughs). If we instead look at 130,000 random people in the database, their final cash sum is $1.3 × 10¹²⁷, or $10 with 127 zeros. That’s roughly the number of atoms in the universe, multiplied by the diameter of the universe in miles, multiplied by the number of milliseconds since the Big Bang. Numbers get big fast when reinvested and compounded unconstrained by earthly friction and liquidity!

So people did way too well. The average stock in the game had returns of 46 percent, which already demonstrates selection bias in my choosing “interesting” levels. But the average player return was 153 percent. It is possible in real life for the average player to beat the average stock, but the figures from the game are skewed because they only count the human players and not the imaginary counterparties to their trades. Finance is not necessarily zero-sum, but the game doesn’t even have anyone taking the other side of player trades, so everybody can buy and everybody can win—cash from nowhere.

Still! It is an illuminating behavioral laboratory, both in comparison to stock markets and in contrast. Based on 10,000 play-throughs of each level (with a total of a million pairs of trades), a lot of looking over people’s shoulders, and some discussions, here is how people played and performed.

As always, the model bears the biases of the modeler; as Wilde wrote, “Every portrait that is painted with feeling is a portrait of the artist, not of the sitter.”

Businessweek economics editor Peter Coy immediately pointed out that the game assumes that, when not invested in stock, your cash is just sitting under a mattress, not being put to any good use. As he played, he didn’t just think about how much cash he was making; he thought about his annualized rate of return. If two li’l green lines representing players’ shares ascend the same amount, but one is steeper, then the steeper line has a higher annualized rate of return. If you got the same percentage return in less time, your cash is locked up for less time, and can be put to good use elsewhere. But the game doesn’t recognize or reward that; it effectively assumes no opportunity cost of capital, because in the game, the only opportunity in the universe is the one single equity in front of you.

That assumption isn’t such a big deal if interest rates are really, really low… which they have been for my entire adulthood. I have grown up basically “discounting” interest. By not considering the carrying cost of capital, I inadvertently made a game by and for the ZIRP generation.

Ultimately, most players will probably just come away with their prior biases confirmed. Some try a few fast trades but then decide that buy-and-hold is the way to go. Others try buy-and-hold, but eventually resolve that, if you want to outperform, it’s crucial to time and capture the swings and bumps along the way. Some people are affirmed in their conviction that markets are basically efficient and active trading is a fool’s game—

Cute @business market-timing game showing how impossible it is, in case you need reminding https://t.co/Pfch3qztaL /cc @jasonzweigwsj

— Paul Kedrosky (@pkedrosky) October 19, 2015

—and others catch on to details like camera movement hints and argue that the actual stock market is similarly rife with “tricks” that can be exploited for abnormal returns.

most realistic thing about new blomberg market game? can't lose money. just like real stocks. on twitter.

— Ezra Rapoport (@HFBondsTrader) October 19, 2015

But what is an “actual” stock market? Are its valuations “actual,” its gains and losses real before realized? Can a company really lose a billion dollars of value in a day? The Trading Game is a coarse map to a territory, but that territory is itself a map, and so on.

Lay not up for yourselves treasures upon earth, where moth and rust doth corrupt, and where thieves break through and steal: But lay up for yourselves treasures in heaven, where neither moth nor rust doth corrupt, and where thieves do not break through nor steal: For where your treasure is, there will your heart be also. —Matthew 6:19–21