The Outside View of Hurricane Dorian
On Saturday, September 1, 2019, when Hurricane Dorian struck the Bahamas, it tied the record for the strongest hurricane on record to make landfall in the Atlantic Basic. Its maximum sustained winds were clocked at 185 mph.
But on the morning of August 27, a mere five days before landfall, the Weather Channel’s official forecast (likely an aggregate of dozens of complex, cutting-edge computer projections called spaghetti models) had Dorian hitting the Bahamas as a tropical storm, with winds around 70 miles per hour.
It was about this time when I started following the storm. Although I’m far from a qualified meteorologist, I’ve been tracking hurricanes ever since I could read an atlas. I used to draw pictures of hurricanes when I was bored at school. And although I had none of the knowledge the computers and meteorologists had—about the upper-level wind shear the storm would face, for example—alarm bells rang in my head the second I saw this forecast. Something was off.
For one, late-August/early-September is the peak of hurricane season. While the hottest times of year on land in the southeastern U.S. are July and August, water takes longer to warm and therefore, the ocean is at its warmest later in the year.
Moreover, the forecast track had Dorian moving through a stretch of ocean called the Gulf Stream, a current of water that moves from the Gulf of Mexico through the Bahamas, up the East Coast, and eventually toward the United Kingdom. The Gulf Stream is part of the reason winters are so mild in the UK; there’s a constant stream of warm water rushing toward it. And that’s the key: the Gulf Stream waters are particularly warm. Hurricanes thrive on warm water, and the warmer, the better.
Further, the forecast called for Dorian to move at an average speed (too fast and a storm can’t properly take advantage of warm water; too slow and it churns up and cools the water beneath it.)
Even its projected path through the Caribbean Sea was favorable for its development: It was unlikely to hit the mountainous middle of the island of Haiti and the Dominican Republic, the only landmass in its path that could weaken the storm.
To me, predicting Dorian to arrive near the Florida coast as just a tropical storm was utterly absurd.
The Weather Channel’s forecast from the afternoon of August 28—about 24 hours later—looked slightly different.
Now, Dorian was forecast to approach the Bahamas and the Florida coast as a Category 2 hurricane, with winds around 100 mph—a definite upgrade from the previous day’s forecast. But I couldn’t believe what I was seeing: They were predicting Dorian would arrive in those warm Gulf Stream waters near the Bahamas as a well-formed, organized hurricane (which is able to use energy more efficiently and therefore strengthen much more rapidly)… but that it wouldn’t actually strengthen much? I couldn’t understand.
The evening of August 28 saw a forecast of a major hurricane approaching the Bahamas … but only barely a major hurricane (Category 3, 111mph-129mph).
There had to be something I wasn’t seeing: wind-shear in the upper levels of the atmosphere, or dry air that would move in and choke the circulation, but as I texted my brother: “A storm…in that particular area…at this time of year…sometimes it’s not so complicated.” In my mind, there was no chance whatsoever that Dorian would fail to at least reach Category 4 status (130mph-156mph).
The rest is history. The forecasts, slowly but surely, increased. Dorian exploded in intensity; on August 31, it became a Category 4, and on September 1, it camped out over Grand Bahama as the strongest landfalling hurricane on record in the Atlantic Basin. Not even I had seen that coming.
Here’s what we can learn: My amateur and unofficial forecasts actually benefitted from being relatively uninformed. I didn’t have wind-shear data (or, really, any other sophisticated data) available to me, and frankly, I was too lazy to look for it.
What I did have was what psychologists Daniel Kahneman and Amos Tversky call “the outside view” in their book Thinking, Fast and Slow.
My instinctive forecasts were based only on a general awareness of historical trends: A storm in that area at that time of year should strengthen, big-time. Sometimes, general knowledge is better than specific knowledge.
I want to make clear: I am not advocating for blindly taking the outside view in every situation. I am not suggesting willful ignorance of reliable data when making predictions. To use the outside view effectively, you actually need quite a bit of intuited data—accrued over years of observation, testing, and evaluating your instincts.
But the outside view is worth taking into consideration as a check on our tendency to be overconfident in our evaluations.
For a quick example, take Cory and Topanga from Boy Meets World.
In the early seasons of the show, Cory thinks Topanga is weird and Topanga is indeed … eccentric, let’s say. At one point she even has a crush on Cory’s older brother. They swear they don’t like each other romantically—they’re even repulsed by each other. It’s not like they’re lying; they truly believe they’re friends and nothing more. But those beliefs are based on the specific information they have, in their own heads, about their own feelings.
Meanwhile, it’s painfully obvious to us outside observers: They’re going to get together eventually.
If you enjoyed reading this — and even if you didn’t — you should check out my book, Ticketless: How Sneaking Into The Super Bowl And Everything Else (Almost) Held My Life Together.