Expected-Value Investing with Low and High Cases
Guess the outcome, halve and double it, then bet if expected value is positive
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 82%
Koch's decision method is deliberately simple. He starts from history's lesson that the future is unknowable, then estimates a single best-guess outcome. Rather than build spreadsheets, he creates a low case by halving the guess and a high case by doubling it, giving a fast scenario band. He estimates the probability of the downside and computes expected value: if an investment could return 100x on success and lose everything on failure, and the failure probability is meaningfully above ten percent but the payoff is large, it can still be worth it, provided it is one position within a diversified portfolio. He pairs this with the wealth-building rule that concentration builds fortunes while diversification later protects them.
Origin
Koch's approach across 37 years of venture investing at roughly 22 percent compounded annual returns; he attributes the scenario habit to studying history and counterfactuals.
Core principles
- 01The future is a land of which there are no maps; treat it as genuinely uncertain
- 02A quick best guess, halved and doubled, beats elaborate spreadsheets for scenario bounds
- 03Positive expected value justifies a bet only inside a diversified portfolio
- 04Concentrate to build wealth, then diversify to protect it
How to run it
- 1
State the best guess
Write down the single most likely outcome for the investment as your central estimate.
Pro tip Do not agonise; a fast honest guess is the input, not a precise forecast.
- 2
Halve and double
Create a low case by halving the best guess and a high case by doubling it to bound the scenarios.
- 3
Estimate downside probability
Assign an honest probability to the failure scenario, consulting more pessimistic people if you tend to optimism.
Pro tip If you are a natural optimist, have a more cautious colleague run the same numbers.
- 4
Compute expected value
Weigh the payoffs by their probabilities to see whether the expected value is positive.
- 5
Check the portfolio condition
Only make the bet if it sits within a diversified portfolio that can absorb a total loss on this position.
Watch out A positive-EV bet can still ruin you if it is your only investment.
In the wild
For Betfair, Koch weighed a scenario where regulation shut it down and he lost everything against a scenario where it returned 100x. Judging the failure probability tolerable and the payoff enormous, and holding it as one position among others, expected value clearly favoured investing.
→ He invested his full spare cash of about 1.5 million pounds and the first tranche returned roughly 100x.
Common mistakes
Over-engineering the forecast
Elaborate spreadsheets create false precision; a quick halved-and-doubled band captures the real uncertainty better.
Making a positive-EV bet with no portfolio
A high-expected-value bet can still be catastrophic if it is your only position and the downside hits.
Is it for you?
Best for
Investors and decision-makers sizing high-variance bets under real uncertainty.
Not ideal for
Single, all-or-nothing decisions where there is no portfolio to absorb a total loss.
From the transcript
“I just say what's my best guess and then I will double the best guess or half the best guess”
“the future is a land of which there are no Maps”
From the episode
#680: Richard Koch — Revisiting the 80/20 Principle, The Power of Optimistic Journaling, Studying History to Improve Investing, and The Grand Beliefs of Winners (Plus: The Toxic Beliefs of Losers)