Thursday links: Bigfoot, re-rolling elections, North Korea, Gen Z investing

Sasquatch science

A study of the Bigfoot-hunting community finds that the search has taken an unexpected turn: “A field that has long been dominated by hunters and outdoorsmen is today also inhabited by a form of ‘citizen scientist’, armed with a pipette rather than a rifle.” 

The authors add that many Bigfooters are “dedicated empiricists who employ methodologies modeled on those found in professional scientific research.”

For an example of the empirical Bigfooter’s mindset, the authors quote the cryptozoologist George Eberhart: “You can apply a scientific method, or a journalistic method to it, to get some good science out of it.”

(I was disappointed to learn that a “cryptozoologist” is not a person who studies crypto enthusiasts, but a person who searches for and studies hidden, legendary, or unverified animals.) 

Bigfooters recognize that their science remains speculative. “There’s plenty of evidence, there’s no proof,” Cliff Barackman told the authors. “There’s a big difference.”

The evidence includes witness testimony, casts of footprints, recorded audio, and even the occasional photograph (the most famous of which is above). 

Some in the community argue that the only conclusive proof that Bigfoot exists would be a body — which makes killing a Bigfoot a scientific necessity.

Most are against the idea, reasoning that since the number of remaining Bigfoot is unknown, killing just one “may doom their population.”

Suspected hair samples have been tested for mysterious DNA. Suspected handprints have been analyzed for distinctive skin oil. “The results have been inconclusive so far.”

“Bigfooting is clearly empirical,” the authors conclude. “We hope readers of this essay, and of our book, appreciate how seriously Bigfooters take this dimension of their work, even in the face of derision.”

The essay is a fun read, but I’d say it’s really about epistemology: It asks how people who diligently gather evidence and conscientiously apply the scientific method can retain their belief in something that seems so ridiculous to everyone else.

The answer, if there is one, would explain a lot about markets, crypto, politics, life…

Prediction markets

Kalshi gave Francesca Hong a 95% chance of winning the Democratic nomination for governor in Wisconsin this week. She narrowly lost.

Kalshi founder Tarek Mansour says the odds weren’t wrong: “A 5% probability doesn't mean it won't happen. It means it should happen 1 in 20 times.”

Does it? Because an election isn’t a roll of a 20-sided Dungeons & Dragons die.

Helpfully, we learned that the Wisconsin electorate was evenly split: Hong won 39.3% of the vote to David Crowley’s 39.8%.

If you could re-run that election 20 times, the electorate would remain close to evenly split every time. 

Hong would not win 19 of those theoretical elections. 

We can therefore say that the correct odds of Hong winning were something much closer to 50%.

Now that the results are in, we have enough information to say that the Kalshi odds were, in fact, wrong. 

There’s no shame in being wrong (thankfully), and in this case especially, as prediction markets are usually well calibrated.

Mansour cites a study from the Washington Post showing that, in the aggregate, prediction market odds get election probabilities about right: Candidates with a 95% chance to win almost always win. Candidates with a 50% probability win about 50% of the time. 

But that’s not the same as saying that every 5% upset was genuinely a 1-in-20 event. Sometimes you really do roll a critical-hit 20. Other times, what the market called a 20-sided die was really a coin flip.

Hong was a coin flip.

North Korea’s finely tuned crypto laundromat

A new study details how North Korea uses a network of launderers, OTC brokers, and peer-to-peer exchanges to cash out stolen crypto.

Much is planned in advance: Pre-arranged wallet structures automatically scatter stolen crypto across hundreds of addresses and venues, including peer-to-peer marketplaces in South Asia or unregulated exchanges in Latin America.

Funds pass through many blockchains along the way, including bitcoin. “There is a belief that the change of ownership often occurs on the bitcoin blockchain,” the authors report. 

The money is typically distributed in increments as small as $30,000, “so a freeze would not be overly impactful.” On peer-to-peer marketplaces, crypto is exchanged for cash in retail-sized increments of around $7,000.

When funds are frozen by an exchange or other service provider, North Korean agents have become newly vocal about getting them unfrozen. Whereas they previously kept a low profile, now, they flood customer service with emails demanding the funds be released — in perfect, but AI-accented English — and with phone calls featuring a lot of shouting and swearing in Korean-accented Chinese.

They also complain loudly on social media, which affects the reputation of the platforms that try to stop them. 

When the crypto is finally converted into physical cash, it’s typically deposited into DPRK-controlled accounts at Chinese banks. 

The system is now highly efficient. In the case of the Bybit exploit, it took just six months for the $1.5 billion in proceeds to be “cashed out and converted into fiat or hard currency.”

You might have a North Korean work colleague 

A Wall Street Journal documentary reports on North Korean IT workers infiltrating the US job market. Not as spies, though. They’re here just to make money. 

The details of the operation are amusingly banal: They use Google Calendar to coordinate work schedules; they quickly copy and paste interview questions into ChatGPT and then read the answers verbatim; they use a US government website to check that the stolen identities they buy on Telegram are legitimate; they have a Discord group. 

Some of it is amusing: An American whose identity was used for North Koreans working nine jobs in nine different states says he’s looking forward to the Social Security benefits they’re earning for him.

Mostly, though, it’s impressive. North Korean IT workers typically work 18- or 20-hour days, often for multiple employers. Some specialize in only doing the interviews. Sometimes they get an American to handle the Zoom calls and split the wages, 50/50.

“North Korea is incredibly innovative,” Wall Street Journal reporter Robert McMillan says. “Their hustle seems to know no end.”

McMillan says the strategy grew out of North Korea’s experience placing workers inside crypto companies. At first, the goal was to compromise IT systems and steal crypto. At some point, though, they seemed to realize the wages they earned were good, too. 

“Hey, we’re getting paychecks for doing this,” McMillan imagines they said to themselves. “Why don’t we just do this everywhere?”

An estimated 90% of the wages go to the North Korean regime, adding up to an estimated $800 million a year.

It’s part of why Pyongyang is kind of booming. Combined, the money earned from IT jobs and crypto exploits added up to about $3 billion last year — nearly 10% of North Korea’s gross national income.

It might also be why getting an IT job in the US is harder than it seems like it should be.

Gen Z investing strategies

A survey from Betterment finds that 26% of Gen Z investors treat sports betting as a deliberate part of their long-term investing strategy. 

52% say they’ve redirected money intended for investing to sports betting at least once over the past year. 14% say they do that multiple times a month.

Only about a third say they don’t bet on sports at all.

Needless to say, this is unfortunate. Sports betting is a negative-sum activity that redistributes wealth from retail gamblers to professional ones, including trading giants like Susquehanna. 

That’s fine as long as people treat it as entertainment. It’s not fine when they believe it’s an investment.

Gen Z investors could learn something from the empirical Bigfooters.

— Byron Gilliam