This is the final exam for Margin of Safety. It pulls together the whole course: the safety factor that engineers build into every bridge and cable, Benjamin Graham’s investor’s discount, sizing the buffer to your own uncertainty and to fat tails, redundancy and fail-safes as margins in disguise, and the trap on the other side — that slack has a cost and a system optimized until none is left is built to snap. Take your time and reason each question through; several look easy until you check the arithmetic or spot the trap.
How this exam works
Read carefully — this exam is final. Each question appears one at a time. Once you submit an answer it is locked for good: there’s no going back, no retry, and no restart. Your score is hidden until the end, where you’ll see a pass/fail verdict. The pass mark is 70%. A few questions ask you to select all correct answers.
What is the cleanest definition of a margin of safety?
Select an answer to continue.
Course Recap
Big picture
Margin of safety, in one picture
- Margin of Safety
- The buffer — where margins come from
- Safety factor is capacity divided by expected load — a factor of 1 is built to fail, so size against uncertainty
- The investor’s discount
- Buy below intrinsic value so a wrong valuation still leaves you whole — protect the downside, since some losses never recover
- Sizing the margin
- Scale the buffer to your error bar and to fat tails, after honestly widening overconfident ranges — no single fixed margin
- Redundancy and fail-safes
- Independent backups multiply reliability, N+1 and slack and defense in depth — but watch for common-mode failure
- The cost of margin
- Slack isn’t free — over-optimization snaps, so right-size to stakes, irreversibility and uncertainty rather than maximize
- The buffer — where margins come from
Key takeaways
A margin of safety is the deliberate buffer between what you expect and what you can survive — built because your estimate is wrong in ways you can’t see. Engineers make it precise with the safety factor (capacity ÷ expected load), and a factor of 1 is a structure built to fail. Graham pointed the same instinct at money: buy value at a discount so a wrong valuation still leaves you whole, and because a 50% loss needs a +100% gain to recover, protecting the downside beats chasing the upside. Size the buffer to how wrong you might be — wider error bars and fatter tails demand fatter margins, and overconfidence quietly thins them all. Redundancy and fail-safes are margins in disguise: independent backups multiply reliability (0.01 × 0.01), as long as no common-mode failure defeats the independence. But slack has a cost, and a system optimized until every buffer is gone — just-in-time, fully levered, 100% utilized — is built to snap. Right-size, don’t maximize: match the margin to the stakes, the irreversibility, and the uncertainty. Build for more than you expect — and know exactly how much more.