Tweet
### Normal

### Fat-tailed

A normal distribution varies a lot in the neighborhood of its average, but produces few examples beyond three standard deviations from that average.

Normal distributions are common in biology. For example, men average 5' 10," and their population has a standard deviation of 4". That means the chance of a man exceeding eight feet (6.5 standard deviations from the average) is astronomically small. And among the billions of men measured, only about 20 have ever stood over eight feet.

A fat-tailed distribution looks normal but the parts far away from the average are thicker, meaning a higher chance of huge deviations.

Fat-tailed distributions are common in society. Since I love documentaries, here's a list of the highest-grossing documentaries. Look how the top three earned dozens of times what any others did. If earnings were distributed normally, these films would be like fifty-foot men. In fact, a single person, Michael Moore, made four of the top ten!

Fat tails don't mean *more* variance; just *different* variance. For a given variance, a higher chance of extreme deviations implies a lower chance of medium ones. To paraphrase Nassim N. Taleb:

*The normal distribution spends 68% of the time within one standard deviation of its mean. If finance has fat tails, how much time do stocks spend within one standard deviation? *

* Everyone answers: 'Less than 68%! Fat tails mean more deviation.' They're wrong: stock prices spend between 78% and 98% of their time within one standard deviation of the mean. *

Both distributions below have standard deviations of 1, but the left is fat-tailed and the right is normal. The slider changes the fat-tailedness of the left (measured here as 'kurtosis') while keeping its standard deviation at 1. As you fatten the tails, the middle bunches up to balance things out.

As each datum is drawn for the deviation plots, its corresponding bar flickers. Look how the fat-tailed deviations stay near the average, but sometimes go above 5 or below -5. The normal is the opposite. The cumulative plots are called 'walks'.

Let's say people deposit their money in your bank, and you use it to place bets. If you think the outcomes of the bets are normal, but they're actually fat-tailed, the bets will still pay off most of the time. But sometimes you'll be very, very wrong. Then the government will have to bail you out to stop a bank run like the one at the end of It's a Wonderful Life .

It isn't just banks that should take notice, though. We also see fat tails in hurricane damage, crop losses, death from deadly conflicts, and other arenas that public policy addresses.

- Wikipedia
- Nassim N. Taleb's in-progress textbook (This visualization was inspired by Vol. 1, Section 9.3.)
- Resources for the Future
- Jun Pan (This is how the distributions are made.)

Made in d3.js by Lewis Lehe and Victor Powell. Lewis Lehe is a PhD student in Transportation Engineering at UC Berkeley, and Victor Powell is a freelance developer/teacher.js.