Tail-Driven Expectations
Plan for eight of ten bets to fail, because two of them carry every outcome that matters
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 85%
The mechanism has two halves that look contradictory and aren't. First, calibrate: in a tail-driven world a few events drive the majority of outcomes, so understand the real base rate — how many startups just as smart, just as funded, just as ambitious failed before you. That number is always very high. Housel's illustration: pick ten stocks and three to five won't exist in ten years, a couple do okay, and one generates all your returns. Start ten businesses, nine fail, one succeeds — that means you did it right. Second, commit anyway: to be a successful entrepreneur you say 'yes, the odds are stacked against me, and I'm still going to plow ahead as hard as I can.' The founder in Housel's story out-pessimised his own investor — 90% chance of failure — and went anyway. The binding constraint on both halves is survival: because failure is the modal outcome, you need a landing pad so a failed bet doesn't leave you destitute and unable to take the next shot.
Origin
Extracted from Young and Profiting. Housel draws it from his own writing on tail-driven outcomes, illustrating with a founder who told an investor the odds of success were 10% not 20%, and Elon Musk's estimate that Tesla and SpaceX each had roughly a 99% chance of failing.
Core principles
- 01A few events drive the majority of outcomes in any tail-driven domain
- 02The path to success often looks a lot like failure
- 03Nine failures and one success is a good outcome, not a bad one
- 04You must know the base rate and plow ahead anyway — that's the founder's dichotomy
- 05Any single bet that can wipe you out ends the game before the tail arrives
How to run it
- 1
Confirm the domain is tail-driven
Check whether a few events really do drive the majority of outcomes here. Housel says it's true of startups, stock picking, and even relationships — you might date ten people before finding your spouse.
Watch out Applying tail logic to a consistency-driven domain just produces avoidable losses.
- 2
Learn the base rate and embrace it
Find out how many people just as smart, with just as much money and ambition, failed before you. Housel says you need to understand that number and embrace it with both hands.
Pro tip Out-pessimise the pessimists on the odds — it costs you nothing and immunises you against the losses.
Watch out Knowing the number is not the same as feeling it; most founders nod and then plan as if they're the exception.
- 3
Commit fully anyway
Hold the dichotomy: the odds are stacked against you and you are still going to give it everything you've got. Housel calls this the contradiction of being a startup founder.
Watch out Half-committing because the odds are bad is the worst of both worlds — you keep the risk and lose the upside.
- 4
Take enough shots
Structure your life so you get repeated independent attempts. A tail cannot land in a single trial.
Pro tip Eight miserable failures and two life-changing wins is a success — count the portfolio, not the last result.
- 5
Build the landing pad first
In the very high odds it doesn't work, make sure you have something to fall back onto — marketable job skills, reserve capital, an un-bet portion of net worth.
Pro tip Run the Regret Test here: most people would regret not starting the business, and would also regret putting in so much that failure leaves them destitute.
Watch out Housel is explicit — a failed business plus no marketable skills to pull yourself back up is a terrible outcome.
- 6
Re-score yourself on the portfolio
Stop reading each individual failure as a verdict on you. Get used to the idea that the path to success often looks a lot like failure.
Watch out Constant disappointment is a feature of the distribution, not evidence you chose wrong.
In the wild
Housel recounts a story he heard years ago from a founder raising money. The investor said: I love the idea of this company, but I think there's only a 20% chance it's going to work — an 80% chance you fail. The founder's reply was that a 20% chance made the investor the ambitious optimist in the room: he himself thought there was only a 10% chance it would work. Housel treats this as the founder mindset in its cleanest form — a 90% chance of failure, fully acknowledged, and total commitment regardless. He notes Elon Musk did the same, estimating roughly a 99% chance Tesla and SpaceX would fail, and going ahead anyway.
→ Calibrated pessimism on the odds coexisting with maximum commitment to the attempt.
Housel applies the same distribution to a stock portfolio: pick ten stocks and there's a very good chance three to five won't exist in ten years, even large companies. A couple will do okay. If you're good and lucky, one will generate all of your returns. He points at twenty years ago, when the biggest stocks in the world included AIG, Enron and General Motors — companies that now either don't exist or barely do. He extends it forward: it's entirely imaginable that one of Amazon, Google, Microsoft, Nvidia or Facebook is gone or a shell in twenty years, and some company nobody is discussing today is worth $10 trillion.
→ You can fail half the time and still do very well, because the tail carries the total.
Common mistakes
Betting everything on one shot
Without a landing pad, the modal outcome — failure — leaves you destitute and unable to take another swing. The tail can only land if you're still in the game.
Reading each failure as a verdict
In a tail-driven domain constant disappointment is the distribution working as designed. Quitting after the losses guarantees you never reach the win.
Knowing the odds and planning as the exception
Most founders nod at the base rate and then build plans that assume it doesn't apply to them. Embracing the number means letting it shape the downside plan, not just the pitch.
Is it for you?
Best for
Founders, investors and anyone making repeated bets where a few outcomes dominate the total.
Not ideal for
Domains with tight, predictable outcome distributions where consistency beats variance.
From the transcript
“The path to success often looks a lot like failure.”
“I know the odds of success are stacked against me and I'm still going to go ahead nonetheless.”
“You want to make sure that you have some landing pad that you're going to fall back onto.”
From the episode
Morgan Housel: How Smart Entrepreneurs and Investors Grow Wealth on Autopilot
Morgan Housel