one life to live · metaphor 1 of 100
Why does advice that is true on average across many people fail for one person living one life in sequence? Because the crowd's fate and your fate are computed with different averages — and in most of the games that matter, the two quietly disagree.
The seminar is persuasive. The opportunity — the leveraged bet, the all-in career gamble, the strategy that worked for the founder on stage — is positive expected value, demonstrably, and the room nods along, because the arithmetic checks out. And yet, of the people who walk out and take it, most end up worse off than if they had stayed home. Nobody lied. The arithmetic still checks out.
The catch is hidden inside the phrase "on average." It quietly meant: averaged across many parallel people at one moment — a thousand of you, each trying it once, their winnings and losses pooled. But you are not a thousand people. You will live one path, in order, and each loss is subtracted not from a pool but from you — from the very capital, health, and credibility you needed for the next attempt. What you lose compounds into what you can no longer try.
Two averages
The ensemble average asks: at round t, pool the wealth of all thousand players and divide. Because heads pays more than tails costs, this number grows a compounding +5% every round — the gold line climbs, exactly as the seminar promised. The time average asks a different question: follow one player through the sequence. Over any long stretch, heads and tails come up about equally often, so the typical round multiplies her wealth by √(1.5·0.6) ≈ 0.949 — a compounding loss. The order of operations is the whole story: averaging then living is not the same as living then averaging.
A world where the two averages agree is called ergodic: there, your long-run experience really does converge to the crowd's statistics, and "on average" is a fair forecast for you. Additive games are like this — win $50 or lose $40 on a fair coin, repeatedly, and your one life does track the ensemble. But multiply instead of add, and ergodicity breaks. The mean rises on the backs of a vanishing few — the one-in-a-thousand path that hit heads twenty times running — while nearly every individual path, including almost certainly yours, decays. The gold line is not a forecast. It is a lottery-winner artifact.
What to try
Press play and watch the gold and blue lines tear apart: the mean climbing, the median — the fate of the person in the middle — sliding relentlessly down. Then hit Reroll "you" ten times. Your coral path jumps to a different player each time, and almost every version of you is losing. Occasionally you'll land on a winner — notice how the win feels like an argument, how instantly plausible "it works" becomes when it happened to you. That feeling is the seminar stage speaking.
Now crank the players to 1000 and the rounds to 300. The gold line keeps climbing — propped up almost entirely by a handful of astronomical outliers — while the "players below start" counter pushes toward 100%. Watch the measured ensemble growth, too: it usually falls short of the theoretical +5%, because even a thousand players is too small a crowd to contain the miracle paths the theory is counting on. Then drop to 10 players and reset a few times: often there is no lottery winner at all, and even the vaunted mean sags into loss. The promise of the ensemble average was always a promise about a crowd large enough to contain a miracle — and you are a crowd of one.
Where lives are non-ergodic
The lens applies wherever outcomes multiply rather than add — wherever a loss shrinks the base from which you attempt the next thing. Wealth is the textbook case: a 40% loss followed by a 40% gain leaves you down 16%, and the money you lost was also the runway for your next venture. But health compounds the same way — one uninsured collapse forecloses the decade of chances that needed a working body. So do trust and reputation: each betrayal multiplies down the credibility with which your next promise is heard.
And beneath all of these waits the extreme case: the absorbing state. Ruin — bankruptcy, the burned bridge, the fatal risk — doesn't set the game back; it ends the sequence. No later round, however favorable, is ever reached. This is why "it's positive expected value" is never, by itself, an answer to "but what if it kills me?": the expectation is computed over parallel worlds, and you do not get to average yourself with the worlds where you no longer exist.
Ergodicity engineering
Humans noticed all this long before the mathematics did, and built machinery for it. Insurance is the purest move: many one-path lives pool their risks, converting each person's multiplicative catastrophe into everyone's small additive fee — the pool experiences the ensemble average that no member could. Community is informal insurance: the family that catches you, the friends who fund the second attempt, restore a reroll to a game that natively offers none. And keeping stakes small — betting a fraction, never the whole — turns a multiplicative gamble back into something nearly additive, so your time average creeps toward the ensemble's. Sizing that fraction has its own mathematics (the Kelly criterion). The principle: don't ask whether the bet is good on average; ask what repeating it does to you.
The mapping
| Mathematics | Life |
|---|---|
| ensemble average | What happens to the average of many parallel lives — the seminar's honest, useless number. |
| time average | What happens to your one life, lived in order — the number that is actually your forecast. |
| multiplicative dynamics | Losses that shrink what you can next attempt: capital, health, credibility as the base of the next round. |
| the highlighted path | No rerolls. You don't get to be the crowd; you get exactly one draw from it. |
| absorbing state (ruin) | Outcomes that end the game rather than set it back — no later round is ever reached. |
| non-ergodicity | When the crowd's statistics are not your forecast, and "on average" answers a question nobody asked. |
Where the metaphor tears
Many small, repeated, independent decisions — which route to work, which restaurant, which week to take holiday — really do average out over a life. For these, the crowd's statistics are your forecast, and treating every choice as a brush with ruin is paranoia, not prudence. This lens is for the large and compounding: the decisions that resize the base from which all later decisions are made.
The coin-flip game lets wealth decay toward zero without limit; real lives are cushioned by family, friends, bankruptcy law, and the welfare state. Every floor and pool partially restores ergodicity — that is precisely what those institutions are for. The metaphor is most literal for whoever has no floor, which is one honest way to define disadvantage: playing the same positive-expected-value game with the non-ergodic version of it.
It is tempting to read all this as "never bet." Wrong average again. Time-average thinking frequently favors boldness — small repeated ventures, cheap experiments, risks with capped downside compound wonderfully along a single path. The instruction is not to shrink; it is to compute the right average before being bold: survive first, so that the long run in "the long run wins" is yours to be present for.