Always buy two new cars

I’ve been driving my wife’s car a lot lately. Her car is nice. It’s smooth, it’s got more room, and it has bells and whistles. It’s always had these things but I’d never noticed.

It’s refreshing to notice instances of relative rather than absolute value. Her car is nice relative to mine but not so nice relative to the newest thing for sale. After driving her car I kinda wanted a new car.

Like made up start ups, the advice to ‘always buy two new cars’ is half a joke. Much of the personal finance advice around here is to choose from pretty good options. Emergency funds should be generally right, both 15 and 30 year mortgages are good choices, and personal finance expertise is from experience not eduction.

To buy two new cars then means that the relative value of the next new car will be largely hidden from me. Sure there will be neighbors and Ubers and advertisements but I make – we make – easy decisions. If it’s not easy to compare then it’s a comparison that won’t occur.


Ironically I noticed this idea with iPhones a couple of years ago. It only mattered that the phone was newer, not that it was newest. All value is perceived value.

Favorites or the field? (part two)

In the first post we jumped off with the idea that the S&P500 is an unbalanced collection of stocks. The top five companies, noted Carl Kawaja, generate 22% of the earnings, so why not only ‘bet on’ the best?

This led to thinking about predictability. Chess is easier to predict than my morning pickleball game. Michael Mauboussin wrote a wonderful book about predicability as framed by skill and luck.

Think about events, Mauboussin explained, as a continuum: from the least skill activities (roulette) to the more skill components (stock markets, hockey, football) to the mostly skill components (basketball, chess). An event’s predictability slides as the skill portion of an outcome increases.

Here is how sport predictions look.
Update two

On the Wharton Moneyball podcast the hosts often talk about betting the favorites or the field. Bettors in events toward the left of the graph are better off betting favorites and events toward the right taking the field.(1)

With two years of data (sorta) it seems that some sports are more predictable (more skill, less luck) than others. Here’s how Mauboussin ranked sports, starting with most skill based: NBA, EPL, MLB, NFL, NHL.

Contrast that with our (sorta) model: EPL, NBA, MLB, NHL, NFL. That’s a pretty good match!

How to use this: to consider if our predictions are more like the NBA or more like the NFL and it’s “any given Sunday” ethos. In NBA-like situations there are a few big issues (players) that drive the results. In NFL-like situations there are more chances for odd bounces (fumbles), subjective decisions (pass interference), or fortuitous circumstances (turnover-worthy pass attempts).

Nothing in real life will be like these sports, but to have a good analogy is a good decision making tool.


(1) “Buying good things can’t be the secret to success in investing,” wrote Howard Marks), “It has to be the price you pay. It’s not what you buy, it’s what you pay. There’s no asset so good it can’t become overpriced.”

The capacity-efficiency tradeoff components

One of the things brought to light by Covid has been the balance between capacity and efficiency. In general, things with high excess capacity aren’t efficient and things with high efficiency lack excess capacity. Another way to think about capacity is as a margin of safety.

Each link in the medical supply chain, wrote Scott Gottlieb, balances capacity and efficiency. Swabs, like those for Covid, are produced efficiently.

“One of the things I learned at the FDA, watching other critical medical products go into shortage, is that it’s often the lowest-margin constituent in a complex supply chain that’s most vulnerable to shortages. Often the only way to produce such a part profitably is to manufacture it at a very large scale, which means it’s likely to be made by a small number of big, consolidated suppliers.” – Scott Gottlieb, Uncontrolled Spread, published 2021

Sixty percent of the swab market was from an Italian supplied, an Italian supplier in the Lombardy region. Eventually the United States scaled up swab productions only to be limited by the machines that ran the tests, another part of the system where the incentives pointed more towards efficiency than capacity. A system is only as strong as the weakest-link-in-the-supply chain.

There is a tradeoff between efficiency and capacity and that tradeoff depends on time and consequences.

“Flatten the curve” demonstrates this. Our highly efficient (low capacity) healthcare system could have been overrun. The consequences of that were huge. One physician friend thought she might have to work in a field hospital – and she’s a dermatologist.

A smaller example is a student taking Algebra. If an Algebra exam is during a sport season, the student will have less time to study being highly efficient with their use of time. The lack of capacity might lead to a lower grade.

However, in the case of Algebra, the time and consequences are more muddled. Is the goal to ‘learn Algebra’ or ‘score 93% on the Algebra Chapter Two Test’? In the case of the former, the time pressure changes.

Personal finance encompasses this idea too. A large capacity is an emergency fund. Those dollars could be invested to earn the historic seven percent. Time here matters too. Typically a person has one source of income (their job) and these vary in pace of payment. A salesperson can work more – and get paid more faster – than a bookkeeper. Their time aspect is different and the consequences for each depends on the difference between income and expenses.

Like the tuning of stereo, these three aspects can be balanced to fit the conditions. There is a trade off between efficiency and capacity. There are consequences in that exchange. And there is cadence for reaction.

Big shocks, like Covid, are out of our control. But though exogenous to our life, there’s still at least three dials to get things that sound right.

History through story

In college I was a campus tour guide and even then had an inkling of the JTBD. While the training consisted of learning dates and statistics, the visitors wanted to know questions about ‘What it was like to be a student here?’.

Historical podcasts offer a similar idea. Rather than precision, it’s appreciation. I don’t remember much from high school or college history, but thanks to historical podcasts, I have some sense of the world. Much like the History through industry books, these are some recent favorites.

1760s The hero of two worlds, The Marquis de Lafayette. My daughters’s favorite character from Hamilton had quite the life experiences, truly a ‘skin in the game’ character. Washington considered him a son, and Lafayette named one of his sons Georges Washington. And it pairs well with…

1790s The Napoleonic Wars featuring Dr. Alexander Mikaberidze. Subtitle, you can’t do just one thing. Following France’s help in the American Revolution was France’s financial debts. This is a strong two hours about the life of Napoleon, his lucky breaks and the world that was kinda set up for someone like him.

1900s Sears Historian Jerry Hancock. Sears is nice to study because it spans a few retail periods: mail delivery catalogs and railroads, urbanization and city center, then stores and suburban malls. This interview is mostly about Sears’s connection to Atlanta.

1939 The start of WWII with Dan Snow. Like the French Revolution, a bunch of factors ‘lined up’. In both cases, one major factor was the economic conditions.

1940s The teenage communist hit squad of the Netherlands. A group of teenage girls are recruited to find (seduce?) inebriated German officers in local bars, convince them to go to the woods, and shoot them in the head. One member, was Hannie Schaft (Wikipedia).

1962 The Cuban Missile Crisis from Dan Carlin. This is one I listen to each October. Also, The Tunnel documentary on a handful of college students digging a tunnel from west to east Germany to help rescue their friends.


Have a suggestions? Let me know!

How to board an airplane?

Everyone knows how to board an airplane, back to front. It’s logical. Back to front means that if someone is taking their time in seat 21C the person in 20A can still seat themselves.

Physicist Jason Steffen built a computer model to see how much faster back to front was relative to front to back. He was shocked. The difference was minimal. Hmm, Steffen scowled at his code, is there a better way?

What if instead of 30 to 1 or 1 to 30, a plane boarded everyone in row 30, then rows 1, 2, 3…. That might work right? The folks in row 30 would have time to stow and seat without holding anyone up.

That kinda works. Steffen’s code continued and compared the 30, 1, 2…29 option to the 1, 2,…30 option. The code noted the faster sequence, and switched around two more numbers. Again it kept the faster option and computed another switcheroo. The fastest boarding process turned out to be boarding every other row.

“It turns the boarding process from a serial process, where one person gets to their place, puts their luggage away, and sits down into a parallel process where you send in fifteen people, say in all the even rows, and they all put their luggage away and sit down at the same time. And then you send in the next group of people.” – Jason Steffen, August 2021

Steffen’s code was a Markov Chain Monte Carlo, a way to solve problems through computer code and exploration. But like wet bias or wait times, ideal solutions may not be the best.

One problem with Steffen’s method is when people travel in groups, especially families. Another obstacle is the culture of air travel, there’s some established norms. Further confounding the case is an airline’s incentives. Faster turns do matter, but relative to upgrades how much does saving time save the company?

Kelly and Crypto

There’s an idea, it’s a formula but really it’s an idea, in gambling called the Kelly Criterion. Broadly, it suggests to act in proportion to edge. Bet big when you have a big advantage. Card counters, like those in the book Bringing Down the House followed this idea.

While Kelly is math, like being Bayesian, it works as a general idea too. Most people never follow the formulaic ‘full Kelly’, rather they bet half or ‘quarter Kelly’ because there’s no way to truly know an edge. So, how exactly does it work as just an idea?

“I’ve had a ton of friends who thought Solana is the future, bought in at a couple of dollars, waited eight months and nothing happened and sold everything. Then, all of a sudden, boom Solana took off. The rapid climb is where a majority of the value capture occurred. You have to build a pretty serious conviction around something and have it be small enough dollars. You can’t say: this didn’t work I’m going to move into the next thing. You have to be able to say: I still have conviction here, I’m going to leave this be.” – Kevin Rose, September 2021

Rose practically uses the Kelly language! Rather than edge and bet he says conviction and small-enough-dollars.

This cost to benefit ratio approach is a nice way to frame decisions. While Kelly started in gambling and moved afield, anything about risk and reward, travel budgets for instance, works.

Most systems have lowish cadences: closer to construction than technology, and the reward portion takes time to compound. When that’s the case, it may help to think about how much conviction we have and how long the cycle may take.


This podcast hit my feed September 19, the same day my wife asked me to buy some Doge Coin. ‘Why’ I asked. I’d convinced her to dollar-cost-average into Bitcoin and Ethereum, but it took a fair bit of convincing. ‘I just want some’ she explained. shrug

Covid and breakfast cereal

cereal selfie

Me, a box of cereal from Aldi, and our kitten. When I bought this cereal the checkout process was 40% faster than rival stores. Plus the cost savings! Sure I had to collect my cart – with the quarter deposit in hand – and also return the cart, but often I meet someone half way and we do the Aldi parking lot exchange of quarter for cart and some goodwill good-to-see-yas.

This Aldi aesthetic is intentional. The cart, the product, the extra long barcodes for the extra fast cashiers are all tactics that support a strategy. I’d heard tactics are not strategy but it’s through the Aldi aisles and Marc Lipsitch’s interviews that the idea becomes as clear and legible as that bar code.

One of the lessons from Covid is how much conditions matter. We’ve learned that individual treatments are dependent on disease stage. We’ve also learned that societal actions are dependent on infection stage. The travel ban, Lipsitch said in May 2020, “was a tactic not a strategy, it was an attempt to show we were doing something rather than a piece of a strategy to make us safe from this virus.”

Strategy is important because our resources are limited. Sure, I’d love to invest in the cryptocurrency of the moment as a lottery ticket but there’s no extra dollars in our investing budget to allocate to a different strategy.

“In the beginning of the intense phase, New York City was working very very hard to do contact tracing at a time when they knew they had lots of cases and didn’t know about most of them. That’s exactly the setting where contact tracing can’t work. No matter how hard you fight the 10% you know about, you’re not doing anything about the rest.” – Marc Lipsitch, May 2020

When tactics fit together like puzzle pieces it creates a beautiful strategic picture. Aldi’s boxes are optimized for speed rather than customer acquisition. Classic cereals use bright colors and cartoons to scream pick me! The cereal aisle is the competition. It’s Cap’n Crunch vs Cinnamon Toast Crunch. Aldi’s products are private labels, so the competition is between big box stores not boxes in stores. The Aldi CAC is speedy checkouts, self-service, and quality goods.

A good strategy has a collection of homeotelic responses. Aldi is one example, but they’ve had almost eighty years to figure it out. Covid is a non-example, but we’ve learned some good lessons and it probably won’t happen again.


There’s a certain amount of what economists call ‘transaction utility’ at Aldi too, we like finding deals. Also, Lipsitch is such a balanced voice on Covid or any field with some uncertainty in the future.

Creativity through randomness

“There’s a great point in your book,” host Adi Wyner says, “where your coach tells you you’re coming in too high in the tournaments.”

“This realization happens pretty early on. It’s six months into my poker playing and I’m very happy I’ve been cashing in poker tournaments. Say you buy in for $100, but the prize money is very heavy up top. Maybe only ten people cash and the person in tenth gets $110. You make a little money on your entry fee, but first place might be $10,000. That’s the disparity. If you consistently cash, but bust out soon after it means you are losing money because it’s not just entry fees, but hotels, travel, non-cashing tournaments.” Maria Konnikova, Wharton Moneyball, June 2020

Konnikova had “settled” on a strategy that seemed okay, but was not. Her coach nudged her off it, framing the true costs of winning poker.

Sometimes a coach knows what to do. Sometimes we are just figuring things out.

“One thing that this pandemic has made us realize is a collective failure of imagination. I’ve been modeling pandemics for twenty years. I’ve been modeling for local agencies, for federal agencies, and all of our models up until 2019 pretty much assumed that our next pandemic was going to be an influenza pandemic. A lot of our planing had been around making sure we had medical counter measures for influenza, playbooks, and game plans. The assumption was that there would be a six-month timeline for vaccines and we would use non-pharmaceutical measures in the interim time period. Covid19 really took us by surprise.” – Lauren Ancel Meyers, University of Texas College of Natural Sciences, March 2021

Covid took us by surprise, in part, because we lacked a coach. In these cases we need creative solutions. Lacking that, random ones work too.

“Each evening during their hunting season, the Naskapi Indians of the Labrador peninsula determined where they would look for game on the next day’s hunt by holding a caribou shoulder bone over the fire. Examining the smoke deposits on the caribou bone, a shaman would read out, for the hunting party, the points of orientation of the next day’s search.” – David Stark, The Sense of Dissonance, August 2011

Stark makes the point that this randomization element meant the tribe would not necessarily return to the last place they succeeded. But it also wasn’t totally random. The hunters still used their very particular set of skills on the hunt.

Lots of life is a balance of “do what I know works” and “find what works next.” The specific mix is contextual, but if we find ourselves in need of a change, and lack a coach, maybe a shaman can point us in the right direction.


Thanks Thomas for the shoulder bone source. Konnikova’s book, The Biggest Bluff chronicles her journey from starter to ‘casher’. We’ve touched on this idea here too: Local maxima of bees, marketers, and NASA engineers.

A landmark numbers walk

We tend to remember more when things are connected. We tend to remember more when there is a story.

For instance, about 1,000 ants equal the length of one beetle. Rather, one Beetle, 1,000 of which lined end to end equal about the length of Central Park. Imagine that. One-thousand VW bugs lined up along the length of Central Park with one-thousand ants lined up along each bug.

Now another step, 1,000 Central Parks is about the width of Australia, ten of which is about the length of the equator.

That’s a nice story and it gives us some Landmark Numbers.

“What I mean by landmark numbers is something that sticks in your brain, you don’t set out to memorize it but it sticks in your brain as a reference point. It becomes a ready reference for you to relate to. These are numbers where, if you have them ready, they help you make yardstick comparisons of the things you hear about. It’s something where you hear a number on the news you go: ‘Hold on.That’s not a big number because it compares to this landmark number in some way.'” – Andrew Elliott, Talks at Google, December 2018

Another landmark number is 4 million. That’s about the number of Americans of any single age. That’s via Tim Harford.

Landmark numbers fit nicely with Maxims for thinking analytically. Richard Zeckhauser suggests that simple cases, extremes, and everyday analogues help us think better. A basic understanding needs context though, and landmark numbers provide just that.


Similar to Landmark Numbers is thinking like Fermi.

Tailing Aaron Rodgers (part three)

It’s time to update two of our ongoing projects. First, to revisit the idea of Aaron Rodgers throwing less than 38.5 touchdowns. The point in parts one and two were to not think specifically about what might happen but think about the states of what could happen. We reasoned there were a lot more things that would influence Rodgers to throw fewer touchdowns rather than more.

Well, no one predicted this. Rodgers now has three games (with the snowy Seahawks showing) with zero touchdowns.

Another idea to update is ‘or the field?’ Are certain events more or less predictive than others? The Wharton Moneyball hosts note that the NCAA Men’s Basketball tournament odds are out.

“We need to do our calculation, how many teams do we have to go down until the odds get to about fifty percent? You probably have to go down six or seven teams, I think I’m still taking the field. I’ll give you Gonzaga, Michigan, Kentucky, Texas, UCLA, and Duke and I’ll take everyone else.” – Eric Bradlow, Wharton Moneyball, November 2021

Well, here’s how the different sports stack up.

November field update

In the EPL and Men’s tennis the top three favorites have a larger cumulative expected odds than NCAA Football and the NBA which have larger cumulative odds than MLB, NFL, or NHL while NCAA basketball seems to be the most unpredictable. The larger the favorites, the thinking goes, the more predictable the skill and the less influential the luck.


Tom Brady continues to chug along contra to the Rodgers reasoning, needing to average 180 passing yards in his remaining games or even just 240 if he misses two. One explanation here is that positive early variance (four games over 375 yards passing) changes things a lot. Prior to this year, Brady only had 23 games with that many yards or more.