Betting on the Tail

The most important decisions in life are bets on tail events—those rare occurrences that seem impossible until they happen, then seem inevitable. Most people understand this intellectually but live as if it weren’t true. They buy lottery tickets while choosing safe careers, chase predictable returns while dreaming of windfalls, and mistake the middle of the distribution for the whole story.

The lottery is the purest expression of tail-betting, which is why it’s so instructive. Everyone knows the expected value is negative. The state takes its cut, the odds are astronomical, and yet people play. The standard explanation—that they’re bad at math—misses the point. They’re not buying a financial instrument; they’re buying permission to dream. For two dollars, you get a week of imagining what you’d do with three hundred million. That’s actually a pretty good deal.

But the lottery teaches the wrong lesson about tail events. It suggests they’re purely random, that you’re either struck by lightning or you’re not. This learned helplessness about extreme outcomes might be the lottery’s real social cost—not the few dollars lost, but the mindset it reinforces.

In business, tail events work differently. You can create them.

Every startup is a bet on a tail event. Not in the sense that success is random, but that the distribution of outcomes is wildly skewed. Most startups fail completely. A few do okay. One in a thousand becomes Google. What makes this different from buying lottery tickets is that founders can influence the odds. Not control them—influence them. The difference matters.

The venture capital industry is built on this insight. VCs aren’t trying to pick winners so much as expose themselves to positive tail events. They know most investments will fail. They’re not looking for companies that might return 2x or 3x. They need the ones that might return 1000x. This sounds greedy until you do the math—without those massive wins, the model doesn’t work.

This creates a strange dynamic where VCs push founders to swing for the fences even when a base hit would be perfectly reasonable. From the founder’s perspective, turning a small profit might be success. From the VC’s perspective, it’s indistinguishable from failure. They’re playing different games with different tail exposures. The smartest investors have figured out how to structure asymmetric bets—limited downside, unlimited upside. Options traders call this convexity. Nassim Taleb calls it antifragility. Whatever you call it, the principle is the same: position yourself to benefit from volatility rather than be destroyed by it.

This brings us to the barbell strategy: put 90% of your assets somewhere super safe and 10% in lottery tickets. Not actual lottery tickets—investments with lottery-like payoffs but better odds. Small bets on startups, cryptocurrencies before they were mainstream, domain names in 1995. Things that might go to zero but could also go up 1000x.

The key insight is that you can’t lose more than 10%, but your upside is theoretically infinite. More importantly, you can make many small bets instead of one big one. The lottery player gets one chance at one jackpot. The barbell investor gets dozens of chances at different jackpots, any one of which could change their life. But the real tail events in life aren’t financial. They’re the chance encounters, the books that change how you think, the conversations that redirect your career. These positive tail events share some characteristics: they’re impossible to predict, they seem obvious in retrospect, and they compound.

That last point is crucial. Financial tail events are usually independent—winning one lottery doesn’t help you win the next. But life’s tail events build on each other. Meeting one interesting person leads to meeting others. Learning one paradigm-shifting idea prepares you to understand the next one. This is why “increasing your surface area for luck” actually works—you’re not just buying more lottery tickets, you’re increasing the odds that each ticket wins.

The danger is that negative tail events compound too. One health crisis can trigger financial problems which cause relationship stress which leads to poor decisions. This is why the most important skill might be avoiding catastrophic downside. Not avoiding all downside—that’s impossible and would mean avoiding all upside too. But avoiding the kind of downside you can’t recover from.

There’s a deeper philosophical point here about how we model the world. We’re taught to think in terms of averages and expect linear relationships. If you work twice as hard, you’ll get twice the results. If you’re twice as smart, you’ll be twice as successful. This mental model works well for many things, but it completely fails to capture how tail events shape outcomes.

In reality, being 10% better at the right thing might make you 1000% more successful. Working twice as hard on the wrong thing gets you nowhere. The difference between a good programmer and a great one isn’t 2x or even 10x—on the right problem, it might be the difference between possible and impossible. This is why career advice that focuses on incremental improvement—get 5% better each year, climb the ladder one rung at a time—can be so misleading. Yes, you need to be competent. But beyond a certain threshold, the returns to incremental improvement flatten while the returns to differentiation explode. The tail events in careers come from doing something different, not something slightly better.

The same principle applies to entire industries. The most profitable companies aren’t usually the ones that are 20% more efficient than their competitors. They’re the ones that are doing something categorically different. They’ve found or created a tail in the distribution of business models. This suggests a different way of thinking about risk. The conventional view is that risk is bad and should be minimized. But if the biggest rewards come from tail events, and tail events are by definition unlikely, then avoiding all risk means avoiding all significant rewards. The goal shouldn’t be to minimize risk but to take the right kinds of risk—ones where the downside is limited but the upside isn’t.

The hard part is that tail events are usually only obvious in retrospect. Before Google, search engines were considered a bad business. Before the iPhone, everyone “knew” that people wanted keyboards on their phones. Before COVID, preparing for a global pandemic seemed like paranoid overthinking. The very nature of tail events is that they exist outside our mental models until they force us to update those models.

This is why the most valuable skill might be what psychologists call “cognitive flexibility”—the ability to update your worldview when reality delivers surprising information. Most people are surprisingly bad at this. They either ignore evidence that contradicts their models or overreact to every anomaly. The sweet spot is maintaining strong views weakly held, being confident enough to act on your current model while humble enough to change it when the evidence demands. There’s a final paradox worth considering. The more you optimize for capturing positive tail events, the more likely you are to experience negative ones too. The entrepreneur who starts multiple companies increases their odds of building something huge, but also of spectacular failure. The investor who makes many small bets on extreme outcomes will have more total losses than someone who buys index funds.

The resolution to this paradox is to realize that not all losses are equal. Losing 100% of a 2% position is manageable. Losing 50% of your entire net worth is devastating. This is why position sizing matters more than picking winners. It’s why the barbell strategy works. It’s why venture capitalists spread their bets across many companies instead of going all-in on their favorite.

The ultimate lesson might be this: in a world shaped by tail events, the biggest risk isn’t taking too many chances—it’s taking too few. The person who never bets on tails is guaranteed to miss both the disasters and the miracles. They’ll have a predictable, stable life. Whether that’s a feature or a bug depends on what you’re optimizing for.

But if you’re reading this, you’re probably not optimizing for predictability. You’re probably wondering how to position yourself for the kinds of tail events that make life interesting. The answer isn’t complicated, though it’s hard to execute: put yourself in positions with limited downside and unlimited upside. Make many small bets rather than a few large ones. Learn to survive the negative tails so you’re still around when the positive ones arrive.

Most importantly, remember that unlike the lottery, you’re not purely at the mercy of chance. You can’t control when or how tail events occur, but you can influence your exposure to them. You can increase your surface area for luck. You can build systems that benefit from volatility instead of being destroyed by it. The house edge in a casino is about 2%. The house edge in life—if you structure your bets correctly—can actually be negative. The universe doesn’t care about taking a cut. Reality offers positive-sum games if you know where to look. The tail events are out there, waiting. The only question is whether you’ll be positioned to catch them when they arrive.