Wars are reshaping borders. Economies are wobbling. The ground beneath a lot of assumptions is shifting. I wanted to share these three books now because they've never felt more relevant. When the world enters uncertain territory, having a framework for making decisions — instead of freezing or reacting — is the difference between navigating the chaos and being consumed by it.
I keep coming back to three ideas whenever life gets messy. Annie Duke's Thinking in Bets taught me to separate good decisions from good outcomes. Nassim Taleb's Antifragile taught me to build a life that gets stronger from chaos. And the concept of power laws taught me that a handful of bets define everything — so cap your downside and leave the upside open.
None of these books told me what to decide. They changed how I decide.
Life is poker, not chess
This is the big idea from Annie Duke. Chess is a game of perfect information. If you lose, you were outplayed. Poker is different — you make the best call you can with incomplete information, and sometimes the cards just don't go your way. Most of life is poker.
Duke's core argument: judge your decisions by the process, not the result. A good decision can produce a bad outcome, and a bad decision can get lucky. Separating those two things changes how you evaluate everything.
Resulting
Duke calls the trap "resulting" — working backwards from an outcome to judge the decision. A deploy goes fine, so the process must have been good. A deploy causes an incident, so someone screwed up. But good processes sometimes produce bad outcomes, and sloppy ones sometimes get lucky. The fix is asking "was the process sound?" not "did it work?"
Thinking in probabilities
Instead of "this will work" or "this will fail," think "I'm 70% confident this will work, and here's what I'll do if it doesn't." You start building contingency plans. You stop being surprised when things go sideways, because you already admitted they might.
The 10-10-10 rule
This one is actually from Suzy Welch, not Duke — but it fits perfectly. How will you feel about this decision in ten minutes, ten months, and ten years? Ten minutes captures the emotional reaction. Ten months captures the medium-term reality. Ten years captures the long game. It's a simple way to zoom out when fear or excitement is clouding your judgement.
Fragile, robust, antifragile
Taleb noticed there was no word for things that get stronger from stress. Fragile things break. Robust things resist. Antifragile things grow. Your immune system is antifragile — small doses of stress make it stronger. A system that fails in small ways, learns, and comes back better is antifragile. It's why chaos engineering and blameless post-mortems matter more than perfect uptime.
The barbell strategy
Be extremely conservative on one side so you can be extremely aggressive on the other. No middle ground. The "medium risk" middle is where most people get quietly destroyed — not safe enough to protect you, not bold enough to pay off. In system design, the boring core protects you while the experimental edge drives growth.
Via negativa
Improve by removing, not adding. Instead of "what should I add?" ask "what should I remove?" Debt. Unnecessary complexity. Noisy alerts nobody reads. The feature nobody uses. The abstraction that makes things harder to understand. The best refactor is often the one where you end up with less code than you started with.
The turkey problem
A turkey gets fed every day. Each day it feels safer. Then Thanksgiving arrives. Stability breeds a false sense of security. The longer things go well, the more convinced you become that they'll always go well — right up until they don't. The people most at risk are the ones who feel safest.
Not all decisions are created equal
Most outcomes follow power laws, not averages. A tiny number of outcomes account for almost all the results. Most startups fail, but a tiny fraction become enormous. Most features you ship won't move the needle, but one might change everything. Some decisions are 100x more important than others.
Capped downside, uncapped upside
If the worst case is small and survivable but the best case is life-changing — that's a bet worth taking. Most people avoid these bets because the downside is visible and the upside is not. Our brains weigh losses more heavily than gains, so we optimise for not losing rather than for winning big. Power law thinking pushes against that. Most bets won't pay off. You only need one or two to work spectacularly.
How the three frameworks work together
These ideas aren't separate for me anymore. They've merged into a single way of thinking about decisions.
Duke taught me to be honest about uncertainty. I can't know the future, so I should judge my decisions by their process, not their outcomes. Think in probabilities. Plan for the world where I'm wrong.
Taleb taught me to structure for chaos. Build a strong, boring foundation — pay off the debt, own the home, keep the core systems reliable — so I can take aggressive swings on the other side. Remove what weakens me. Stay suspicious of prolonged calm.
Power laws taught me which swings to take. Look for asymmetry — capped downside, uncapped upside. Accept that most bets won't pay off. Stay in the game long enough for the ones that do.
The sequencing matters. You don't take the big swing first and hope the safety net appears later. You build the net, then swing. That's why I paid off the mortgage before trying the startup. That's why I saved before stepping away from my role. Each boring, conservative move expanded the menu of bold moves I could afford to take.
Not every bet paid off this year. But the ones that did — the career, the family, the move — more than made up for the ones that didn't. And that only worked because no single loss could take me out of the game.
In uncertain times — and these are uncertain times — that's the only strategy I trust. Build the net. Stay in the game. Swing when the asymmetry is right.