Watch

Listen

I write a post each morning. By lunch, that post is a video on YouTube.

My writing was already running. Multi, the tool that renders videos, was already running. The video is the joint output. Writing produces the post that goes in. Multi turns it into the video that comes out.

A flywheel is a system where each part feeds the next. Mine has many parts:

• my writing
• Multi
• the YouTube channel
• paying customers

Each part is a node. A tap is a single action that touches more than one node at once.

When my morning post becomes a video, my writing gets a second life on YouTube. Multi gets a tag in the video so viewers learn the tool exists. The video also stress-tests Multi so I see what to improve next. The YouTube channel gets one more video in the library.

One video. Three nodes move at once.

Some viewers see the tag, visit Multi and become customers. Their money funds more videos. More videos earn more viewers. More viewers means more customers.

The sum is bigger than the parts because one input becomes many.

Most actions you take do one thing. A flywheel turns one action into many moves.

♻️

What action are you already taking that could be feeding a second node if you let it?

Watch

Listen

One line from my prediction markets video kept looping in my head. The market price was supposed to reflect what the crowd believed. It was reflecting those beliefs. It was also changing them.

I’d read about something like this years ago. Reflexivity. George Soros, the British pound, 1992. He shorted ten billion in pounds. The bet predicted the pound’s collapse and helped trigger it. He made over a billion dollars in a single day.

Two posts came out of that. The first named Soros and the mechanism. The second covered the people who can move prediction market outcomes. Both stayed in finance.

Belief works the same way in your life. When you act on what you believe, people see the action and react.

• Opportunities open or close.
• Results come in.
• Sometimes the ones you wanted. Sometimes others.

Each result shapes your next move. The loop runs in the direction of your “bet.” Refuse to bet and the loop stays at zero.

Psychologists named one direction learned helplessness: failures, real or imagined, train you to stop trying. The internet named the other manifesting. Same process, opposite directions.

Soros acted on his belief and the pound fell.

When other people refuse you, you have feedback. You can reflect and adjust. When you refuse yourself, you make zero progress.

🔄

Where have you been rejecting yourself before anyone else could?

Watch

Listen

I spent two days on a name and shipped zero code. Domain searches across a dozen tabs. No progress on the thing that mattered.

Then I stopped. Deployed on a subdomain. Started shipping.

The name will come. A product with users and revenue pulls the right name toward it.

Over 5 years ago I started publishing weekly reports on my personal site. Once readers kept coming back, I registered the trends.vc domain.

The naming decision feels productive. It feels like building. But you’re naming a thing no one has used yet.

Now I deploy on a subdomain before I have a name. The decision is reversible. I can keep shipping. If users keep coming back, the naming decision gets easier. If they don’t, I don’t waste more time on it.

Naming is one version of the larger trap.

• You compare tech stacks for weeks before shipping
• You design a logo before the landing page exists
• You debate a podcast format for weeks before recording the first episode

Each is a delay disguised as preparation.

🚀

What reversible decision are you delaying?

Watch

Listen

I knew I was going to open Instagram before I touched the screen. The sleep cost and regret were clear. I opened it anyway.

That’s the dual-use problem. We use the same technology to learn languages, fix our squat, study a market and scroll at 2am. A useless app gives you a clean no. A tool that helps without taking much gives you a clean yes. Instagram helps and hurts at the same time.

• Mass media: The news helps you understand the world. It can also be propaganda that programs you.
• Algorithms: A feed teaches you a craft in a month. The same feed takes a year of your evenings.
• AI models: An AI model writes a clear paragraph for you and writes ten thousand more for spammers.

The pattern holds at every scale. Nuclear physics gives power plants and bombs from the same equations.

The capability and the danger are the same thing.

I deleted my Instagram account. For YouTube, I use Unhook, a browser extension that strips the recommendation feed. The tutorials load. The recommendations disappear.

The useful technology is also the dangerous one.

🔥

What’s one technology worth cutting or constraining this week?

Watch

Listen

Drag costs you more than speed helps.

I spend two days a week on email and support. I want to cut that to one. The instinct is to push harder. Bigger engine. More fuel. Hire help.

I’m thinking like an F1 engineer instead. There are two ways to make a car faster: add power or remove weight. The math says removing weight wins.

Performance is power divided by weight. Start with 100 over 100. Ratio: 1.

Apply the same 50% move to either side:

• Add 50% to power. Ratio is 1.5.
• Remove 50% of weight instead. Ratio is 2.

Same percentage move. Removing weight wins by a third.

Run it again.

• Add another 50% to power. Ratio is 2.25.
• Remove another 50% of weight. Ratio is 4.

Each removal of weight matters more than the last. Each addition of power gives the same lift it always did. Removing weight compounds. Adding power doesn’t.

The same shape shows up in loss aversion. You start with $100. You lose 50%. You have $50. You gain 50% back. You have $75. The same percentage that took you down can’t bring you back up.

Same math, both directions. Losses compound against you. Removing weight compounds for you.

This is also why I pick removal experiments first. Cut a feature, a chapter, a meeting. The downside is bounded. I can add it back if I cut wrong. The upside compounds. Less to maintain. More energy for what’s left.

The math favors removal.

🪶

Where are you adding power when you should be removing weight?

Watch

Listen

I have been studying investors lately. Reading how they think, how they bet, how they hold. One pattern keeps surfacing.

The best ones think evolutionarily, whether they say so or not.

Dalio is the most explicit about it. The frame I keep coming back to is r vs. K selection.

Y Combinator funds 200+ companies each batch. Most won’t reach profitability. A handful hit escape velocity. The expectation is a power law: a few investments return everything and more, so the rest can go to zero.

That’s r-selection. In biology, r-strategists maximize offspring count at low investment per individual. Rabbits. Bacteria. Most won’t survive, but enough will to carry the line forward.

Warren Buffett runs the opposite system. Berkshire Hathaway holds roughly 50 positions, but five carry most of the value. He buys companies he plans to hold forever. The filter is strict. The conviction is deep. Each position earns its slot through a rigorous process before capital moves.

That’s K-selection. K-strategists produce few offspring and invest heavily in each one. Elephants. Great apes. High investment, high survival rate, high ceiling per individual.

The instinct is to pick a side. Both are adapted to different environments.

YC works because the market is the unknown. Even great teams lose when the market is wrong. The only way to find which markets win is to fund broadly and let results decide. The batch is the filter.

Buffett works because the signal is strong. He operates on decades of pattern recognition. The companies he evaluates have:

• track records
• moats
• management teams

Narrow selection is rational when input quality is high.

The strategy follows the environment.

r-selection when signal is weak and options are cheap. K-selection when signal is strong and depth is what creates the edge.

I run my own work this way. I start many small experiments at once. The ones that show results get my full attention.

Each makes the other possible.

The failure mode is applying r-selection to a situation that needs K, or K to a situation that needs r. Running hundreds of experiments when you already know what works. Or picking one bet before the data tells you which one.

The question is which environment you’re in.

🧬

Based on your current environment, which strategy should you run?

Watch

Listen

I want a word for this.

There’s a quality I keep seeing in habits, systems and businesses that last. Something beyond hurdle rate. Hurdle rate asks, “Is the input big enough to count?” Tardigradity asks, “What are the odds it keeps going?”

A 5K every morning that ends in week two cleared the hurdle and went to zero. A push-up every Sunday for a year is too little input to build strength. The first one dies. The second one falls behind. Compounding needs both.

Tardigrades survive radiation, vacuum, boiling, freezing. Five mass extinctions. So do sharks, horseshoe crabs, nautiluses, sponges and ostracods. The pattern: survivors are small, slow, low-metabolism, flexible in diet, tolerant of weird conditions. They won by asking the least of the world.

Tardigradity is the probability a thing keeps going.

Every requirement a habit has is a place it can break. Speed, intensity, equipment and willpower each add to the list. A tardigradic practice keeps the list short.

A gym habit needs:

• Motivation
• Time
• A drive
• Working knees
• Clean laundry
• A gym that’s open

Six things can stop it. A push-up in the kitchen needs one: you, for ten seconds. Same muscle group. Radically different tardigradity.

Intensity is seductive. It depends on peak conditions you can’t guarantee. Tardigradity looks like weakness because it asks for so little. That’s what makes it survive your worst weeks, your traveling months, your sick years.

Ask if the habit is ambitious? Wrong question. Ask if it clears the hurdle. Ask if it’s tardigradic.

🐢

Which of your habits would still hold during your roughest stretch?

Watch

Listen

I have a backlog of 200+ rough blog post ideas. When I sit down to write today’s post, the idea is already there. I captured it weeks or months ago.

Coming up with an idea is one kind of work. Writing the post is another. Doing both in the same session is context switching.

With a backlog, I separate the two jobs. Ideas land whenever they land. I write on my own schedule. One job at a time.

The same habit works beyond writing. Problems are business ideas. When you run into one, write it down. Now you have a backlog of business ideas ready when you need one. Running into problems starts feeling good.

💡

What are you figuring out and finishing in one sitting?

Watch

Listen

I built five books in one day to pick which one to ship.

For 2,000 days, I ran a daily founder standup. 44,110 standups in total. I needed to turn it into a closing artifact. The data was rich. Picking the angle was the hard part.

I could’ve picked the one that felt right and committed. Instead, I built five parallel versions. One book per angle:

1. First Sale to Full-Time (milestone tracking through the logs)
2. The 2,000 Day Arc (my personal journey through my own entries)
3. Building Wisdom from 2,000 Standups (distilled principles)
4. The Longest Standup (the whole thing as one continuous narrative)
5. The Morning Pages of Founders (Julia Cameron framing)

Each got a full outline, a cover concept and a pitch. Enough weight that I could feel what each book would be like if it existed.

Number five won.

Once all five existed on paper, one of them was clearly harder to vary. Each sentence in the Morning Pages framing carried weight. The others had load-bearing slots I couldn’t fill without flinching.

This is simulation thinking: render a working version of each option, then pick the one that holds up when you can compare them.

Most decisions stall because the options are still descriptions. “Book about milestones” sounds good. “Book about wisdom” sounds good. Each description sounds good until you render it.

The fix is to render enough of each option to compare them. Five is usually the right number. Three is the floor. One is what people do by default, which is why decisions feel so stuck.

Before AI, outlining five books took five days. AI can render all five in an hour. Now you can run simulation thinking on any decision that matters.

The counter-instinct is that this feels wasteful. Four books you’ll never write. Hours spent on versions that die. But the alternative is committing to the wrong angle and finding out ten chapters in.

You can’t judge a book by comparing it to four blank slots.

🎭

What stuck decision would unstick if you simulated five futures instead of imagining one?

Watch

Listen

I lived in a new city every few months for 5 years. Each place meant rebuilding from scratch. Where to eat. How to get around. Where to work. Which grocery store. Which gym. Which cafe had decent wifi.

I thought the hard part was logistics. Packing. Flights. Apartments. The real cost was invisible. Each new city handed me thousands of decisions I didn’t know I was making.

Each environment has a surface area of decisions it forces on you.

A new city has a massive decision surface. Which neighborhood is safe. Which route to the grocery store. Which grocery store. What’s the tipping norm. How does the transit work. Where’s a good place to focus. Can I walk here at night.

A place you’ve lived for two years has almost none.

You eat at the same five places. You know the route. The gym is the gym. The grocery store is the grocery store. Currency is currency. These aren’t decisions anymore. They’re defaults.

A stable environment turns whole categories of decisions into defaults before you face them.

• A walkable neighborhood makes transportation a default.
• A stocked kitchen makes food a default.
• A separate bedroom makes “where do I work” a default.
• A familiar city makes navigation, language, currency and logistics all defaults.

People say “routine helps productivity.” True, but it undersells the mechanism.

A routine is a stack of pre-made decisions. When you wake up and do the same thing each morning, you aren’t exercising discipline. You’re skipping decisions. The discipline was front-loaded: you made the choice once. Each morning after that, the choice is already made.

Environment design works the same way, one level deeper. A stable environment makes the decisions beneath your routines automatic.

Deciding to exercise is easy when the gym is two minutes away. Picking where to eat is quick when I have three places I love. Choosing a route is automatic when it’s muscle memory.

You spend less willpower because you make fewer decisions.

Five years of nomading gave me something valuable: data on what environment variables matter. Light. Walkability. People. Quiet. A separate workspace. I wouldn’t have that list without the exploration.

But the exploration had a cost I couldn’t see while I was paying it. Each new city reset my defaults to zero. Each month in a new place meant hundreds of decisions that a settled person handles on autopilot.

The productivity gains from adapting to your environment are real. They’re about decision count. Fewer decisions. More energy. More output.

🪫

What’s one decision you make each day that your environment could turn into a default?