Book Review: On The Edge, by Nate Silver

Nate Silver published On The Edge: The Art of Risking Everything on August 13th and, well, I knew I had to jump it on the top of my reading list. 

On_The_Edge

Silver is best known for creating the election-forecasting blog FiveThirtyEight, which he reconstituted on Substack in 2024 as The Silver Bulletin

For newspaper-reading political obsessives like me, it’s Nate Silver’s world between now and November 5th, and we’re all just living in it. He aggregates and weights polling data, inputs other calibrated factors into his model, and suggests a probabilistic approach to electoral college results. 

In his new book, Silver gives coherence – and the name “The River”  – to a community of practitioners and thinkers who I admire, and who I try to channel in my own writing. 

Silver’s thesis is that there is a particularly successful and newly salient group of people from a variety of walks of life who share a common epistemology. 

Epistemology is the 50-cent word for a theory of knowledge or a way of thinking. 

The River

So this group, the River, primarily shares a way of thinking about risk. 

Among other things, they think probabilistically about risks in rational and objective ways, rather than emotional or traditional ways. They compare the probability of success versus the size of the rewards. They specifically seek to take risks when there is positive expected value – where the size of the reward is big enough to overcompensate for the probability of failure. They are competitive. They are strategically empathetic, by which he means they try to see how the other side of a contest is thinking. They update their views when new data comes in. They try to not be overly wedded to one world-view or one model of how things work. They can be contrarian in the face of societal consensus. 

The meat of the book is derived from interviews and observations of people who share this epistemology from the worlds of technology, private equity, trading, gambling, cryptocurrency and artificial intelligence. Silver is a member of The River, so he’s eager to explain the advantages of this method, as it serves him and others well when investing, gambling, sports-betting, election-model building or other risky endeavors. He’s also a careful journalist and writer, so he sifts through – especially in later chapters – how this type of thought can go wrong for individuals or the world.

Nate Silver

For an example of the latter, you get in this book very close-up conversations with Sam Bankman-Fried before, during and after his spectacular cryptocurrency rise with FTX, and his subsequent fall and fraud conviction

You also get the most in-depth explanation of the rise of artificial intelligence I have read to date, including an attempt at a technical explanation of how large language models (LLMs) like ChatGPT work for a non-technical reader like myself. You’ll get far more poker history and lore and strategy than you’ll ever need, as well as the methods and thought process of a sports better.

My Own Retrospective Guide

Narrative, connectivity, identity, justice, and status-quo pattern recognition, are examples of other useful intellectual techniques common in academia, government, and journalism. They also may be at odds with the hyper-rational, probabilistic, contrarian risk-orientation of The River.

What I hadn’t expected is that Silver’s new book would provide a kind of retrospective guide to my own mental aspirations when writing this column. I naturally gravitate to stories about practitioners from The River, probably because I think it’s a great corrective to the typical epistemology of traditional journalism. 

While there are quite a few members of The River and an extensive philosophical tradition – as explained in detail throughout On The Edge – the vast majority of us do not apply enough of these thought processes.

Silver dedicates two chapters to the rise of artificial intelligence, and especially the worries of leading rationalists like Eliezer Yudkowsky who see an existential threat from AI, something I became alarmed about last year

Silver is a major advocate for prediction markets like Manifold, Polymarket and PredictIt, which allow the collective bets of crowds on outcomes in a probabilistic manner, and with which I’ve become obsessed in the past few months.

The River’s way of thinking informs my view of why retail options trading is not likely to be profitable in the long run. 

My views on the organization GiveDirectly – which attempts to bring a rational and probabilistic mindset to philanthropy – stems from this same impulse. 

The Recent Texas Lotto Example

This one didn’t come from Silver’s book, but an excellent Hearst investigation of a lottery scheme in Texas is one of the best recent examples of River vs. non-River thinking from the Lone Star State.

The most commonplace piece of personal finance wisdom is to never buy lottery tickets. And this is true, you should not, precisely because the “expected value” of every lottery ticket you’ve ever bought is less than the price you pay. The more tickets you buy, the more you will lose over time, like any other game of chance at the casino. This is Expected Value 101.

On the other hand, if there were a theoretical lottery game in which the payout had a positive expected value, then you should play that lotto. In the real world this is extremely rare, and requires specific circumstances and some sophisticated techniques.

The investors and implementers behind a lottery scheme in 2023 are an example of people from The River who know how to calculate and exploit expected value opportunities, even with lottery tickets. You should look up the Hearst investigation yourself as its quite interesting, but the short version is this: 

For the April 22, 2023 Texas Lotto drawing, an investment group managed to spend an estimated $25.8 million to purchase every numerical combination possible in order to guarantee a win of the $57.8 million lump sum offered by the Texas Lotto, plus smaller prizes as well. Their expected value calculation depended upon the payout getting very large over many months without a jackpot, plus their confidence they had solved the technological and logistical problem of buying up every number combination over the course of two days. They basically brought an Oceans 11 approach to winning the Texas Lotto, and it was all legal. 

If you don’t know how to do that, you should not ever be buying lottery tickets. 

As a p.s. to the story, the Texas Lottery Commission will probably change the rules to prevent this kind of exploit in the future.

Improve Our Thinking

I’m not claiming to be particularly great at The River’s mode of thinking, but I am naturally attracted to it. I aspire to it.

My interest began as a childhood board game player, was enhanced by years working on Wall Street, and is kept percolating through hobbies like dabbling in poker, investing, and prediction markets. 

I’ve been exposed enough to it throughout the years to see it as something that can give me, and other people, a possible edge in understanding the world. Whether you’re a member of The River already, or just want to avoid the pitfalls of conventional thinking, I recommend Nate Silver as your guide.

A version of this post ran in the San Antonio Express-News and Houston Chronicle.

Post read (18) times.

The Best Presidential Economic Policies

Last week I mentioned the worst economic proposal of the candidates, so in contrast today I’ll mention the best. 

Democratic presidential nominee Kamala Harris’ plan to support the building of 3 million new housing units is it. 

YIMBYism and Texas

One of my strongest held economic growth views, solidified over the last 15 years in Texas, is about housing.

My theory is entirely too reductive, but I have come to believe that the key to Texas’ astonishing economic strength is housing, and specifically the ease with which it can be built. This keeps housing affordable relative to other places, which attracts newcomers, which in turn spurs growth, in a virtuous cycle of economic development. 

By contrast, I moved here from New York City 15 years ago. A big reason why housing is so expensive in New York, and California, and my hometown in Massachusetts, is that it’s too darn difficult to build new housing. There is something about all the zoning, the regulations, the permitting, and the scope for local objections that all adds up. A culture of NIMBY (Not In My Back Yard), expressed by both local government and by neighbors able to block new projects, prevails. The local administrative state raises the cost of new construction to the point where housing is chronically underbuilt in those areas. Which leads to unaffordable housing.

The backlash movement to this chronic problem, whose worldview I have come to adopt, is known as YIMBY (Yes In My Back Yard). 

Texas, in contrast to these high-cost-of-housing places, seems highly capable of increasing housing supply as the demand increases. I think the state’s YIMBY approach is really working here. Sometimes (again, I know this is too simple, but I like simple) I think efficient home building is the single greatest comparative advantage of Texas versus slower-growth places. 

Harris’ housing policy

So that’s the background to my excitement over Democratic candidate Harris’ policy announcement this month of a plan to encourage 3 million more housing units to be built. 

Harris
She got this one right

As a former Democratic Senator from California, the epicenter of NIMBY, Harris was not necessarily an expected advocate for YIMBY zoning reform. Which, from a messaging perspective, might make it even more important. Democrats at the national level seem to be getting the message.

In 2022, the Biden administration announced a federal focus on affordable housing as a mechanism for making housing more, well, affordable. 

Reading that laundry list of 2022 “pro-housing” policies feels extremely on-brand for traditional Democratic thinking, which is to say, the proposal and implementation of 33 distinct well-meaning, micro-targeted programs across the federal bureaucracy and throughout different parts of the country, each of which might do something positive but none of which trust anything could happen without creating a new governmental carrot or stick to make that very specific thing happen.

[Note to Ric/Editor: I know that previous sentence is ridiculously run-on, but it’s my attempt at having the sentence style match the content point, if that makes sense?]

Policies like new incentive grants to rehab outdated affordable housing. Help for rural counties to build multi-family units. Promotion of R&D in modular housing. Finalization of rules about “income averages.” The eyes glaze over. I hope that all did something positive? I don’t know.

The Harris announcement in August 2024 is much better. 

It starts with the correct big idea, which puts “building more homes at the center of their economic agenda because rents are lower and homes are more affordable when we build more housing.” 

Yes.

Compared to 2022, the Harris campaign proposes a far more markets-based solution to building millions more housing units. Their new message is a classic YIMBY agenda: lower the costs and time to build by dropping zoning, permitting, and review processes. If you increase the supply of the thing (Yes! Wow! Amazing!) prices stay low. Also, how to do this? 

Specifically, her campaign proposes to incentivize regional and local authorities to lower many of the barriers to building. Exempt projects from very slow environmental review. Change HUD codes to expand the types of houses that can be built. Streamline permitting to speed up construction. Accelerate historical preservation processes that often trip up rehabilitation of properties. 

Each of these is thematically linked to pro-growth YIMBYism and just actually better than the 2022 ideas. It’s about getting the regulatory state out of the way to allow the construction sector to do its thing, faster and cheaper.

Housing policy wonks have responded very positively to the Harris proposal, and are somewhat shocked that it has become a key plank of her campaign.

Trump, by contrast, is no YIMBY. He has made “protecting suburbs from apartment buildings” a talking point of his campaign, which is a classic NIMBY move. Trump is on the wrong side of this, choosing to use fear-based language aimed at suburban voters. The way you “protect suburbs from apartment buildings” is you impose NIMBY regulations. 

yimby

Directionally, the GOP has been committed to the idea that the government is best when it does the least, which should have a YIMBY flavor to it when applied to local zoning and permitting rules around new construction and property rehabilitation. So you might have expected the Trump campaign to be better on this than the Harris campaign. But it’s not.

To be clear, I don’t endorse the view that local government regulation has to be minimized in all or even most situations. I’m just saying that in the case of housing we have a notion about what smaller-government GOP plans ought to be, or could be. It’s what has been going on in GOP-led Texas. The Texas model – pro-growth, pro-development, light regulatory-touch, light zoning – gives us a hint of what could be. 

But as far as I can tell, Trump has no “build baby build!” plans when it comes to housing in America, just a nod at NIMBYism.

Harris’ centering of this pro-growth housing approach has the potential to signal to Democratic-led areas – like California, New York, and my hometown in Massachusetts – a better way.

It’s not like we have lost the knowledge or willingness to build housing. It’s just that we have allowed a NIMBY set of policies to get in the way of lower-cost housing and therefore national economic well-being. 

A version of this ran in the San Antonio Express News and Houston Chronicle

Please see related posts:

The 2024 Presidential Candidates’ worst economic ideas

Post read (8) times.