Updating My Priors: Short Takes 9/2025

Top forecaster Peter Wildeford was interviewed and explained his heuristics:

• Follow a "nothing ever happens" approach, betting against dramatic predictions since most provocative forecasts don't materialize; people overestimate the likelihood of exciting but unlikely events.
• Rather than relying solely on his own analysis, he built a network of domain experts across different topics and aggregates their insights; his edge comes from having good sources rather than original analysis.
• "For example, when Israel attacked Iran, I know three people with good track records on Middle East issues. I talk to all of them and try to draw the through line. To be frank, I add very little of my own value in my forecasting. I just make out like a bandit by having these people I trust and drawing good averages across them."
• Embrace humility: realizing there's a lot he doesn't know and learning when to bow out.

The last bit reminds me of a line by Nassim Taleb that it is important to surround yourself with books you have not read to remind yourself of how much you do not know.

---

Marshall McLuhan was a Canadian philosopher and media theorist best known for pioneering the study of media theory and coining the phrase "the medium is the message" to highlight how the form of communication (not just its content) profoundly shapes society. 

He developed a concept of information overload that centers on the idea that the advent of electronic media introduced a state where individuals are constantly exposed to more information than they can process. 

McLuhan observed that "one of the effects of living with electric information is that we live habitually in a state of information overload. There’s always more than you can cope with". 

He argued that the constant bombardment of stimuli and information through media leads not only to overwhelm but also fundamentally alters how people perceive, process, and engage with the world.

When we are faced with information overload,' he taught; 'the mind must resort to pattern recognition to achieve understanding' seeking to make sense of chaotic, fragmented information by identifying simple narratives or overarching pattern.

It reminded me of the Kafka/Lynch study I reference in Agency Detection and the Confident Pundit where people got better at a pattern recognition game after reading Kafka or watching a David Lynch film, since the content is so bizarre as to appear meaningless.

---

Nate Silver on when prediction markets are reliable.

There’s not a super simple answer to that question, but here are some factors I’d consider:

• How liquid is the market?
• How sophisticated is the modal trader in having domain knowledge about the subject at hand?
• Whether or not the average trader knows what he’s doing, is there a class of professional traders with the capital to correct any imbalances?
• Is the situation easy to model through statistical techniques?
• Is it a repeated event where traders have the opportunity to refine their estimates through trial and error?
• And is there local knowledge — or even insider knowledge — that might contribute to the wisdom of crowds?

---

In the New York Times, David Brooks writes:

“The anti-institutionalists have advantages. It’s much easier to degrade and destroy than to preserve and reform.”

This reminds me of a James Lindsay analogy about the seduction of being in a demolition crew. It doesn't take much to tear something down; much harder to build. I have little faith that those who want to tear down institutions can rebuild something better. And now the right has fallen to the same folly as the postliberal left. They have become anti-builders.

---

In Masculinity and Dignity, I wrote about the 60/40 rule: In any given group, when the gender ratio exceeds 60% women, men start to leave and it becomes harder to recruit new men. However, there is at least one exception: nursing.

From The Wall Street Journal: 

The number of male registered nurses has increased from about 140,000 in 2000 to about 400,000 in 2023. This means that about 14% of nurses are now men, up from about 9% roughly two decades ago.

But …

About 46% of nurse anesthetists, who administer anesthesia and help patients recover from it, are men. The job currently requires at least a master’s degree, and the average salary is $214,000.

At the other end, only about 10% of certified nursing assistants are men.

… so men might be going into specializations that are below 60% women. This seems to track with Scott Alexander’s finding that although doctors have achieved gender parity, the specialities men and women go into track with gender stereotypes; with men doing systems-based things like surgery and women doing people-based things like pediatricians. 

It's the End of Democracy As We Know It (And I feel fine)

 I know I said I was done with posts about Trump, but this is actually a post about people's reaction to Trump. 

Yes, you can see the paralells between his appointments of loyalists and the chilling of speech from his executive orders, and how those paralells match historical examples of dictators overthrowing governments. 

But how likely is that to happen?

The best way to answer that question is to not just rely on my judgement or on the thoughful articulation of some writer who already agrees with all of my opinons. That's right, baby. Time for some prediction markets!

First off: we need something specific and measurable to know if change has occurred. The global democracy index seems well suited for this. The Economist recently ran an article about this. They have the U.S. listed as a "flawed democracy" with a score of 7.85. (Full report here.) A hybrid regime would be less than 5 and an authoritarian regime would be less than 4.

Kalshi, which is real people betting real money, asks "How much will US democracy weaken under Trump?" They give an 86% chance of being below 7.80, so it's unlikely to go up, and a 31% chance of going below 7. There aren't any markets below that, so the people with skin in the game seem to be betting that it is unlikely we are anything below a flawed democracy. 

Next up, Metaculus. Not money being exchanged, just superforecasters chasing after bragging rights. However, they have shown to have the most calibrated and accurate forecasts of all the prediction markets.  When I search on their site I can find the question "Will the United States drop below a 7 on the democracy index by 2040?" but when I click on it I get an error message. Now, that's a long time line and there are only 37 forecasters (but given their historical accuracy, I don't ding them too many points for that sample size) but the consensus is 40%. Higher than I am comfortable with but still seems unlikely for us to fall into an authortarian regime. 

Finally, Manifold. Like Metaculus, no actual money. But unlike them, anyone can throw up a precition market. This gives us lots of questions to explore but spreads out the sample size a bit. 

First up, with 191 traders: "If Trump is elected, will the US still be a liberal democracy at the end of his term?" At the time of the election this was 80%. It has now dropped to 58%.

With 39 traders: "Will the democracy score for the United States significantly decrease by the end of 2025?". On Feb. 4, this was 28%. It is now 76%. This uses a different ranking, which has the U.S. at 0.892 and defines a decrease as sliding down in score by 0.1.

With just 11 traders we are using another index that has the U.S. at 83/100. "If Trump wins, what will the Democracy Index be for the US at the end of his presidency?" Only 25% of traders think it will be below 78, but even a score of 68 would still put us in their highest category of "consolidated democracies".

Trading at a sample size of 21, "If Trump wins, will the US electoral democracy index fall below 0.80 by 2025?" They reference Our World in Data, which uses the V-DEM index and has us at 0.85. They give it a 91% chance. I don't know how to put that number in context, but they note "0.80 is the lowest central estimate for the US since 1975."

Lastly, with 277 traders, "Will Trump 2.0 be the end of Democracy as we know it?" They define this as "Will there be a major change with Trump's election such that the U.S. does not live in a traditional US Democracy anymore? Such conditions would include things like: Trump running for a 3rd term, declaration of Martial Law at any point, removing Supreme Court Justices or ignoring a material order from SCOTUS, Executive Branch overturning a certified election result, refusing a transition of power, etc."

Finally, a bold stance! Anyway, the traders have converged on a 33% of this happening.

The Upshot: first, we should put more stock in what these markets tells us than what your wacky aunt posts on Facebook or your favorite cable news host rants about every evening. Second, it seems that our democracy ranking is likely to decrease but not at a level that we should be worrying about, like our country being overthrown by MAGA heads. 

But things can change. I look back to revisiting this post and these markets over the next four years to know how much I should be updating my priors. 

My Final Trump Post

Source: https://www.flickr.com/photos/rickfoster/8760051427

I recently watched Mr. McMahon, the Netflix documentary on WWE titan Vince McMahon. It was entertaining on several levels but the one part that stuck was the section on Donald Trump. Vince kinda sorta took credit for Trump’s ascendency to the White House. Not through any sort of cross promotion or campaign donation or election strategy. Vince’s claim, which I believe, is that Trump learned from WWE wrestlers how to be a performer and then appropriated that to his public image.

It is something that now feels blindingly obvious in retrospect. The boasting, the narcissism, feeding off both the audience’s adulation and the scorn of one’s enemies—these are all traits that Vince and Triple H and Steve Austin have been honing for decades. 

Wrestling legends like Ric Flair built their characters around ostentatious displays of wealth, unrelenting self-promotion, and a refusal to show humility. Trump employed the same persona in his campaign, boasting about his wealth, intelligence, and business success, much like Flair’s infamous catchphrase: “To be the man, you gotta beat the man.”

Pay attention to their blatant rule-breaking. In wrestling, heels cheat openly—whether through eye pokes, low blows, or interference from an ally—only to then revel in their lack of consequences.

Heels thrive on eliciting a reaction. Whether being showered with boos or ironically cheered, they control the crowd by feeding off their emotions. Trump’s campaign rallies functioned much like WWE events, complete with entrance music, taunting slogans (“Lock her up!” “Fake news!”), and call-and-response interactions with the audience.

I recently read a quote from a swing voter who had decided to support Trump, saying “At least he’s an honest liar.” Out of context, that sentence makes no sense. And yet, I know exactly what he means.

I found professional wrestling to be more entertaining once I learned it was all fake. You become more focused on the personalities and how they perform for the crowd. With Trump, most of his supporters know he is lying when he gets up on stage and makes outrageous claims. But they know this is part of the act, and they find it entertaining enough to put up with all the corruptiony stuff that gets reported on in The New York Times.

Yes, the WWE heel is not a good person and all of the fans know that. You know who else isn’t good? The arena’s security guards who keep pulling fans out of the stands for holding up offensive homemade signs, escorting them out of the building, and putting them in cuffs. Only they are not a part of the show. And when the heel, in full character, promises to demolish those security guards, whose side will the fans be on?

This is where Trump feels authentic to his voters. All his lies appear to be on the surface, so they let their guard down. The so-called "honest liar" allows them to feel vulnerable in comparison to the traditional politician who appears to be lying with a straight face, and the median voter feels unable to get a read on them.

When I think of all of the presidents of my lifetime, I can surmise a reason that compelled them to run for the executive office. Some feel overregulation has harmed small businesses, others feel inequality has made it near impossible for people to succeed without a robust welfare system. But for the life of me, I cannot think of a consequentialist reason that Trump wants to be president, what he gets out of it. I think he simply likes being president, all the pomp and circumstance and attention.

So here is my goal for the next 4 years: I am not going to give him the thing he wants, my attention. If Ezra Klein does a 3 hour podcast episode about what to expect from Trump's first term I will gladly skip over that and wait for him to interview someone about AI. If The Atlantic runs a column about how broad tariffs or DEI crackdowns are terrible, I’m not going to read those.

Inevitably, I will hear about some policies he has passed and what they have done. But I don’t care and don’t want to know about the latest outrageous thing he says. He is not worthy of my attention or frustration. The paraphrase Adam Serwer, the attention is the point. Stop giving him what he wants.