Sunday, August 9, 2015

The wisdom (and foolishness) of crowds

In 1906, a British smart guy named Francis Galton attended a fair and there was a contest where people would try to guess the weight of an ox. Being a statistician (and therefore presumably single), he wondered how the crowd as a whole would do, and he figured that since most of them were not experts in cow-related matters (no, their wives don't count), they'd probably come up with something laughably wrong.

The ox ended up weighing 1,198 lbs.  The average of the answers the crowd came up with was 1,197 lbs, which means that the collective guess of those amateurs was only a pound off.


NPR recently replicated the experiment and uploaded a photo to the internet, asking people to guess the weight of a cow named Penelope.  Over 17,000 people submitted guesses.  The actual cow weighed 1,355 lbs, and the average guess was 1,287 lbs.  Not as close as the original, but still less than 5% off.  Of the respondents, 3,000 were "experts," people who knew a little something about cows, and their average was 1,272.  That's right, the self-proclaimed experts actually did worse than people who knew nothing about cows.  Not by much, but still.

While at the fair, they asked a group of kids how much they thought the cow weighed, and after the first one said 200 pounds (because kids are stupid), the others followed suit and gave answers that were pretty close.  Now, I'm tempted to attribute that to the stupidity of children as well, but adults do that too.  It's what's called "information cascade," and it happens because people instinctively look to leaders.  They assume since someone already blurted out an answer with some confidence, they must know what they're talking about.

There were a handful of people who guessed the exact weight, and the guys at NPR called up one of them at random, who turned out to be a 20 year old college student, who came up with the answer via the very scientific method of "googling that shit."  His extensive 90-second research told him that the average weight of a cow was 1500 lbs, and by comparing the provided picture of Your Mom with the picture of Penelope, he correctly guessed she weighed 1355.

The two points I'm making here are that the crowd tends to do better than the experts at predicting things, as long as they aren't swayed by the decisions of those around them (like the kids were, by that first little dummy), and that just because someone gets the right answer once, it doesn't mean they will be able to do it again.

In the same vein, if left to its own devices, the stock market tends to be a very good indicator of the actual value of a stock.  Investors are irrational, but collectively, they tend to arrive at a price that's pretty close to what it's actually worth, as long as they aren't swayed by "experts" or TV personalities who talk about this or that stock or investment.

That's also why it's so hard to beat the stock market.  The wisdom of the crowd is pretty good, and the people who beat it are often just lucky.  Remember the cow experts who did worse than the average non-expert at guessing the weight?  Financial experts like money managers tend to do the same with investments.  More often than not, they actually underperform.  Four out of five do worse than the market over a 20-year time period.


Now, the principle of crowd wisdom doesn't really hold when a certain investment is highly visible and when investors are influenced by others.  When people see everyone is piling into a certain stock, they think that it must be a good investment, so they do as well.  Then bubbles and panics happen:

Remember bitcoin?  The real estate market? Fuckin' Beanie Babies?
Chances are, once you hear about something hot, it's probably too late.  That was the case with bitcoin, the real estate market, you name it.  By the time the media reports on it, the smart people have made their money and will soon be pulling it out.  Of course, there is that 20% of experts and money managers who actually do better than the market, and those genius investors like Warren Buffett and Peter Lynch, who have 9- or 10-figure net worths, but good luck with that.  Your best bet as a novice investor (as I would assume most of the people reading this to be) is to ignore the chatter and not try to go for whatever the hot investment is at the moment.  Instead, you'll be better off by investing regularly in a low-fee index fund.  To read more about cows, crowds and averages, check out this article.

No comments:

Post a Comment