Learn
Risk-adjusted decisions weigh the downside as carefully as the upside
A risk-adjusted decision compares options on what they are likely to return and on how badly they could go wrong, rather than on the hopeful case alone. Finance has worked this way for decades. Any choice with a serious downside benefits from the same discipline.
The idea in finance
In 1966 the economist William Sharpe proposed measuring a fund's return above a safe rate against the variability of that return, so that a fund earning 10 percent with wild swings could be compared fairly with one earning 8 percent steadily. The Sharpe ratio, as it became known, rests on a plain point: two returns of the same size are not worth the same if one carries far more risk.
Three figures for each risk
Outside finance, a workable method needs three figures for each serious risk: how likely it is, what it would cost if it happened, and how much of that cost remains after the mitigation you will actually fund. Multiply the likelihood by the remaining cost to get the expected loss, then subtract that from the option's expected gain.
Expected loss is not the whole story. A risk that would end the business deserves weight beyond its average, because there is no recovering from it, however unlikely it may be.
Applying the method
List the options and the benefit each is expected to bring. For each, name the two or three risks that could change the ranking, and estimate the three figures for every one. Adjust the benefit, then look separately at the worst single case. Choose the option with the best adjusted result whose worst case the organization could survive.
The common mistake
Listing every conceivable risk so the analysis looks thorough. Forty risks with no numbers produce a long document and no adjustment at all. The opposite mistake is just as common: counting a mitigation at full value because it appears in a vendor's brochure or a policy, when nobody has funded or tested it.
Risk on the record
In a Decision Room, the Gray Wolf hunts hidden weaknesses in each option, the Red Teamer runs stress scenarios that could break the plan, and the Grandfather separates what can be undone from what cannot. The engine works the expected values from your own figures. The Judge's Verdict states which option leads once the downside is counted, what it rests on, and what would change it. The Human Decision, including how much risk to accept, belongs to you.
Two suppliers, level on average
Supplier A saves $300,000 a year over the current supplier but runs a single factory, with a 15 percent chance in any year of a three-month outage costing about $900,000 in lost sales. Supplier B saves $180,000 and runs two plants, with a 5 percent chance of a smaller disruption costing $300,000. Adjusted, A is worth $300,000 less 0.15 × $900,000, or $165,000 a year. B is worth $180,000 less 0.05 × $300,000, also $165,000. On average the two are level. A's worst case is three times larger, so a firm with thin cash reserves should prefer B, while a firm that could absorb a $900,000 loss might reasonably take either.
Questions
- What does risk-adjusted mean in plain terms?
- Judging an option by what it is likely to return after subtracting what it could plausibly lose, rather than by its best case.
- How do I estimate the likelihood of a risk?
- Start from history where you have it: how often similar suppliers, projects, or markets have failed. Where you have none, give a range and check whether the decision changes across it.
Further reading
Bring Argumont a Decision worth challenging.
Start with a paid 60 day pilot on your own real Decisions.