Margin of Safety

I don't need to be right about the future to invest wisely — I just need enough room between what I pay and what I believe something is worth to survive being wrong.

• Calculate margin of safety using the formula (Estimated Value − Price) ÷ Estimated Value on a real numeric example.
• Explain why an estimate of intrinsic value is always somewhat wrong, and why the buffer size matters more than the precision of the estimate.
• Distinguish price from estimated value, and identify the common mistake of measuring the discount against price instead of against value.
• Decide, given a price and an estimated value, whether the gap between them is wide enough to act on.

Here's the core problem: any estimate of what a company is really worth is a judgment call, not a fact. You can be careful, use good data, and still be somewhat wrong — because the future is uncertain, and nobody can measure it exactly. So the real question isn't "how do I get a more precise number?" It's "what do I do knowing my number is probably off?"

The answer is margin of safety: the gap you deliberately leave between the price you pay and your estimate of value. It's calculated as (Estimated Value − Price) ÷ Estimated Value. Notice the formula divides by the estimated value, not the price — value is the reference point you're protecting, and price is just what the market happens to be asking today.

Margin of safety isn't there to make your estimate more accurate. It's there to absorb the error that's already baked into any estimate. If you buy something at a price very close to your estimate of its value, you have almost no room — the smallest overestimate turns into a loss. If you insist on paying well below your estimate, that gap becomes a cushion. When your estimate turns out to be too optimistic (and often it will be, at least a little), the cushion is what stands between you and losing money.

The buffer, not the precision, is what actually protects you.

This is why two people can land on nearly the same value estimate and have completely different outcomes — not because one of them calculated better, but because one of them left more room for being wrong.

So the buffer, not the precision, is what actually protects you.

How much buffer is "enough" isn't a fixed number. It depends on how confident you are in your own estimate and how much error you can afford to absorb. Someone estimating the value of a stable, well-understood business might be comfortable with a smaller gap; someone estimating something harder to predict should demand a wider one. What matters is deciding your required buffer in advance — before a price looks tempting — so you're not negotiating with yourself in the moment.

Rohan and Kavya both spend an evening independently working out what they think a company's shares are really worth. They land on almost the same number: ₹100 per share.

Rohan checks the market price the next morning — ₹95 — and buys immediately. It's only ₹5 below his estimate, but it still looks like a deal, so he doesn't think twice.

Kavya looks at the same ₹95 price and decides to wait. She's not confident enough in her ₹100 estimate to act on such a thin gap. She sets a target: she'll only buy if the price falls to ₹70, giving her a wide cushion between price and her estimate of value. Weeks later, the price does fall to ₹70, and she buys.

A few months on, new information comes out about the company — weaker demand, higher costs than expected — and analysts revise the fair value down to about ₹80. Both Rohan and Kavya had overestimated the company by the same amount; their original judgment was equally off.

But the outcomes are very different. Rohan paid ₹95 for something now worth ₹80 — he's sitting on a loss. Kavya paid ₹70 for the same thing — she's still fine, even a little ahead.

Run the formula on each: Rohan's margin of safety was (100 − 95) ÷ 100 = 5%. Kavya's was (100 − 70) ÷ 100 = 30%. Kavya wasn't a better analyst than Rohan — her estimate was just as wrong. She was simply protected by a wider gap. That gap, not her judgment, is what let her survive being wrong.

Point: The size of the margin of safety — not the accuracy of the original estimate — determines whether a decision survives being wrong, and the calculation must always be measured against the estimated value, not the price paid.

Let's stretch the Rohan-and-Kavya example a little further to see why buffer size is a personal decision, not a universal rule. Suppose a third investor, Meera, also estimates the same company at ₹100, but she requires a much stricter buffer — she'll only buy at ₹60 or below, a 40% margin of safety. She ends up not buying at all, because the price never drops that far before new information arrives. Was she wrong to miss out? Not necessarily. Meera simply decided in advance that she needed more room than Kavya did, likely because she trusted her own estimate less, or because she couldn't afford to be badly wrong on this particular decision. Rohan's 5% buffer failed him. Kavya's 30% buffer protected her. Meera's 40% requirement protected her too, at the cost of a missed opportunity. None of these buffers were "correct" in some absolute sense — each reflected how much error that investor could tolerate. That's the real decision margin of safety asks you to make: not "what's the right number," but "how much room do I personally need before I'm willing to act?"

• Believing that if you research more carefully or build a more detailed calculation, you eventually reach the 'true' value and no longer need a margin of safety. — Effort feels like it should produce certainty — the more work you put into an estimate, the more correct it feels. Fix: Remember that even careful, well-researched estimates remain approximate. The margin of safety exists precisely because no amount of analysis makes a future-facing estimate certain.
• Assuming a calculated margin of safety means you're protected from any loss. — The word 'safety' sounds absolute, and a clean formula feels authoritative, so it's easy to read it as a guarantee. Fix: Treat margin of safety as damage reduction, not damage elimination. It lowers the chance and size of a loss if your estimate is wrong — it does not promise a profit or rule out a loss entirely.
• Calculating the discount against the price paid instead of against the estimated value — for example, treating '₹30 off a ₹100 estimate' the same as '30% off the price.' — Everyday shopping discounts are usually expressed as a percentage off the price, so that habit carries over by mistake. Fix: Always divide by the estimated value: (Estimated Value − Price) ÷ Estimated Value. Value is the reference point you're protecting, not the price you're paying.

• Margin of safety is the gap between what you pay and what you believe something is worth: (Estimated Value − Price) ÷ Estimated Value.
• Every value estimate is somewhat wrong — the goal isn't a perfect number, it's enough buffer to survive an imperfect one.
• The size of the buffer, not the accuracy of the estimate, is what determines whether a decision holds up when you're wrong.
• Margin of safety reduces the damage of being wrong; it never guarantees a profit or rules out a loss.
• Always measure the gap against the estimated value, not against the price — value is what you're protecting.

Q1. What is the correct formula for margin of safety?

• (Estimated Value − Price) ÷ Estimated Value
• (Estimated Value − Price) ÷ Price
• (Price − Estimated Value) ÷ Price
• Price ÷ Estimated Value Answer: (Estimated Value − Price) ÷ Estimated Value — Margin of safety is measured against the estimated value, since value is the reference point you're trying to protect — not the price you happen to pay.

Q2. True or False: If you research a company carefully enough, your estimate of its value becomes certain, so you no longer need a margin of safety. Answer: False — No amount of careful research removes uncertainty about the future. Even a well-researched estimate is still a judgment call, which is exactly why a buffer is needed.

Q3. Two investors estimate a company's value at almost the same number, but one leaves a much wider gap between the price they pay and that estimate. When the estimate turns out to be too optimistic, why does the investor with the wider gap fare better?

• Their estimate of value was more accurate than the other investor's
• The wider gap absorbed the error in the estimate, even though both estimates were similarly wrong
• They had access to better information about the company
• Margin of safety guaranteed their investment would be profitable Answer: The wider gap absorbed the error in the estimate, even though both estimates were similarly wrong — The buffer, not the precision of the estimate, is what protects a decision. A wider gap between price and estimated value leaves more room to absorb being wrong.

Q4. An investor estimates a company's shares are worth ₹200 each. The current market price is ₹150. What is the margin of safety? Answer: 25% — Using (Estimated Value − Price) ÷ Estimated Value: (200 − 150) ÷ 200 = 50 ÷ 200 = 0.25, or 25%. Notice the calculation divides by the estimated value, not the price.

Q5. A stock is estimated to be worth ₹1,000 and trades at ₹700. One learner calculates (1,000−700)/700 = 43%. Another calculates (1,000−700)/1,000 = 30%. Which method does this unit actually teach, and why does the mismatch matter? Reveal: Weak: doesn't matter, both are roughly the same gap. Strong: dividing by Price (43%) vs. Value (30%) are genuinely different numbers — the taught method divides by Value, since value is what you're protecting; dividing by Price overstates the real buffer.