How to blow $1 million (when cryptocurrency meets hubris)

A common argument is that rich people can buy their way to success. I have argued that commentators, on either side of the aisle, overestimate the importance or conversion rate of money into influence. Money is necessary for tangible things, such as houses and cars. It’s not so great for buying a legacy, intellectual credibility, or influence on discourse. Either you you vast sums of money, or it does not work. And any influence money does have, tends to be very fleeting.

On March 17th, 2023, former Coinbase CTO and crypto evangelist Balaji Srinivasan wagered $1-2 [0] million on Twitter with a friend that the price of Bitcoin would be at least $1 million in three months. The bet was set up in such a way as to ensure Balaji’s loss, and was locked in a ‘smart contract’ to ensure transparency. I don’t know how much he is worth, but given he has millions to throw around on a whim, presumably a lot.

On May 2nd, for reasons that are not clear, Balaji closed his bet early instead of waiting until the full 90 days had passed, awarding his friend the $1 million. That same day on his official website he published an article, “The Fiat Crisis“, capitalizing on the media coverage of his loss, to explain his thesis in more detail and his rationale for giving away $1 million. That was his intent: to spend $1 million, effectively as an advertisement, to prove a point about the unsustainability of America’s debt burden, and also as an endorsement of Bitcoin. The stunt went viral, but the follow-up article didn’t. The few times his article was shared, like on Reddit and Hacker News, it didn’t go viral, got few votes, and was panned in the comments [1].

For example, the Reddit link above shows his article was shared just once. Yes, one whole share after spending $1 million. (If you ever feel like a failure, just remind yourself of this.) And three shares on Hacker News. By comparison, articles by Bari Weiss, owner and founder of, not uncommonly get many shares, and without her having to spend a million dollars, such as the recent article, “Is Justice Still Blind in Canada?” which was shared at least ten times on Reddit (although it was written by contributor Rupa Subramanya, not Ms. Weiss). This show how far money goes, which is not very.

But how about Twitter? Because he made the bet on Twitter, and his Twitter account has 920k followers, so surely it must gone viral? But we’re talking about the article, which is what is what he paid for indirectly. Not the bet itself going viral, but the actual end-product, that being the article. According to a Twitter search (you have to be logged in to see this because of stupid changes by Elon), his article was shared a grand total of 25 times, so $60,000/share if we go by the $1.5 million figure in the footnote. Ouch. He could have just paid top Twitter influencers far less to post it (even the most expensive influencers only charge $10k/post). Also, the most recent share was on the 8th of May, so it’s already forgotten. By comparison, that article about Canada was shared way more times despite having only been published on the 8th and not spending $1 million to promote it. He probably could have gotten more mileage had he written the article first, kept the million, and used his VC and other connections to publicize it. I am sure he could have found at least 25 followers who would have shared the article on their Twitter timelines for free. [2]

I don’t need to belabor the point about why he’s wrong about cryptocurrency or the national debt. I am more interested in why he failed so badly. Even if he is wrong about crypto, there is little denying he is successful by almost any objective measure. He’s worth millions, maybe even billions. He was an executive of a multi-billion dollar company. He has nearly a million Twitter followers, including followers in high places, such as people in VC. So why did his stunt fail so spectacularly?

I posit the necessary confluence of luck and skill to succeed at punditry or as a thought leader are scarcer than being an executive, and is not something you can put a price tag on, which all but assured his failure. Yes, becoming successful pundit is harder than even being an executive of a billion-dollar company, both in terms of rarity and the skills involved, which I will expound on in an upcoming post. The pointy-hair boss meme is not entirely wrong: executives do a lot of delegating, and even if they fail are still well-compensated.

Regarding punditry, there are three major hurdles: getting the facts/details right, empathy/likability, and gauging sentiment. The audience for writing is far more astute and fickle than just shareholders, laypeople, or other employees. Look how long it took for blatant frauds like FTX, Enron, or Theranos to be uncovered, yet readers are so good at finding errors in writing and punching holes in arguments. In business, if something works you can hone in on that. The ‘business world’ tends to rewards specialists whereas punditry rewards generalists, the latter which is possibly harder because a much greater breadth of knowledge is required. Also, he made no effort at explaining or addressing the obvious counterexample, that being the failure of Bitcoin to hedge inflation in 2022, when it mattered most, as discussed above. In 2022 CPI gained the most in decades, about 7%, yet Bitcoin fell about 65%, for a real lost of over 70%…terrible.

Worse yet, even if you’re right, it may not be good enough because reader tastes can change on a dime. I have encountered this myself numerous times where arguments I had made pre-2021 are no longer well received today, even controlling for audience. Thus a certain amount of ‘mind reading’ is necessary. There is a whole art to having good opinions, which is a skill in itself. It’s deceptively easy (most people fail). Someone like Joe Rogan makes it look effortless. He’s never misread his audience once in his 13-year podcasting career. He’s always on top of things, which is remarkable. Even the best CEOs cannot maintain such a long streak. Even Trump misread his audience and was booed after he said he was vaccinated. Delivery matters too: good verbal intelligence and eloquence is also needed (PowerPoint and ‘corpspeak’ will not cut it).

Finally, there is empathy and likability. Even if you get the other two right, appearing unlikable will doom your case. I think this is why his stunt failed, for the same reason Romney’s $10,000 bet, which he made half-jokingly, also backfired (and you have to mess up pretty bad if the media is taking the side of Newt Gingrich). It’s a major distraction, which is why his stunt went viral but his article didn’t, having been overshadowed by the gimmick. All people could talk about was how stupid the bet was or his rapacity, not his ‘fiat crisis’ thesis.

People in STEM sometimes like to downplay the journalism profession as easy “an AI can do it” or “learn to code”. To that I say, go ahead and try it. [3] Maybe you can do better than Balaji and not need to spend $1 million (it’s a low bar). Overall, people in business like to think they can reform Washington. Surely, having all that private sector experience must be an asset. Maybe it is, but Trump 2016 notwithstanding, it’s not something where outsiders can just waltz in and turn over the tables of discourse or policy (and even Trump and his libertarian advisor Peter Thiel, collectively worth billions, were stymied in their efforts to reform government). It does not work that way.

[0] Balaji sent $2 million USDC, which is a type of stable coin, to a smart contract. But $1.5 million appears to have been paid out, but the title card says $1 million:

[1] If you truly believe inflation is going to surge, why waste your time and money guessing or hoping Bitcoin will go up (which it certainly won’t). Just take out a long-duration fixed-rate loan. People who took out 30-year fixed-rate mortgages in 2020 and early 2021 made out like bandits. People who bought Bitcoin in 2021 lost anywhere from 40-75%. The evidence from 2022 and 2020 clearly shows Bitcoin is not a hedge against crisis or inflation, and is just another highly correlated risky asset that benefits from ‘easy money’, similar to stocks. Even cash holds up better. During Covid, Bitcoin lost 30% of its value in a single day, and lost 70% of its value in 2022 as inflation was surging. He’s also conflating credit risk with duration risk. Long duration bonds got clobbered because rising inflation expectations have an outsized effect on the long-end of the curve. He also says that insurance companies, which hold bonds on their portfolio, did badly, yet insurance stocks held up well in 2022:

[2] He claims his article was read 162.5 million times:

But even if you just think of the bet as a type of public service announcement, it easily ratio’d the New York Times by a factor of 30X+. Sulzberger spends $170M+/month ($2.05B/year) for 604.6M pageviews in March 2023, whereas this PSA cost a mere $1.5M in a month for 162.5M direct pageviews. Citizen journalism defeats corporate journalism once again!

Rubbish. Maybe he means 162 million Twitter impressions on his tweets about his bet, but there is no way his article was viewed by anywhere close to that number of actual humans. Especially given the link was only shared 25 times on Twitter, the most recent share being ten days ago. And does a rich guy spending $1 million to spread FUD about the financial system in order to boost his Bitcoin investment, really count as citizen journalism or someone who has your best interests in mind?

[3] On the SATs and GREs, the verbal section is known for being harder than the math–so much so the notoriously hard analogies section was removed from the SATs in 2005–so I never got this notion of verbal being ‘easy’.