跳到主要內容
Apexhone 想透徹

Investment & Finance

Margin of Safety Thinking

安全邊際思考 · Source: Benjamin Graham / Warren Buffett

Any decision resting on critical assumptions — investments, ventures, major commitments — ensuring you survive when assumptions prove wrong

Core Concept

If your critical assumptions are off by 30%, does the decision still hold? Margin of safety isn't pessimism — it's acknowledging forecast uncertainty and building in enough buffer so you're not eliminated when things go wrong.

When to use this

For any decision with potential irreversible loss: investment sizing, startup cash runway, career-change savings buffer, health-related risk exposure. Mandatory when failure cost dwarfs success upside.

When not to use this

Adding a margin to low-risk, high-frequency choices forfeits compounding — e.g. small daily life calls. In highly competitive arenas, over-conservatism also loses to those who shoulder risk.

Questions you will be asked

Using this framework, you will work through —

  1. 1.What decision are you considering?
  2. 2.What three critical assumptions must hold for this decision to work?
  3. 3.If each assumption is off by 20-30%, does the decision still hold?
  4. …and 3 more

Worked example

Expand to see what a filled-in run looks like

Situation

考慮辭職創業,預計 18 個月後產品能獲利。家裡 24 個月生活費。

1. What decision are you considering?

辭職創業,計畫 18 個月達到月入打平。

2. What three critical assumptions must hold for this decision to work?

1) 18 個月達到打平;2) 我能單人完成 80% 的開發;3) 早期 100 個付費客戶可在 6 個月內取得。

3. If each assumption is off by 20-30%, does the decision still hold?

若假設都偏差 30%:時程變 23 個月、開發要找 1 人外包(額外成本 60 萬)、客戶獲取 8 個月。預算會超過家裡的 24 個月生活費。

4. Where is your margin of safety? How bad do things have to get before you regret this decision?

安全邊際在第 21 個月。如果到 21 個月還沒打平,我必須回去找工作——這時履歷有 21 個月空白,求職較困難。

5. Under a "things didn't go well but within reason" scenario (not extreme), can you absorb it?

不確定。如果 18 個月只達到打平的 50%,我心理可能撐得住但配偶不會;財務上會吃到緊急預備金。

6. How do you need to adjust this decision to have sufficient margin of safety?

調整:1) 把目標改為 12 個月達到「能負擔自己生活費」(不是公司打平);2) 同步保留 1 個遠端兼職收入;3) 設定第 12 個月為 review 點,沒達標就部分復職。

Use it inside ChatGPT / Claude

Paste the prompt below and the AI will walk you through this framework, one question at a time.

你現在是引導使用者做「安全邊際」評估的決策教練(Graham / Buffett)。
依序問:
1) 你正在考慮的決策是什麼?
2) 這個決策成立需要哪 3 個關鍵假設為真?
3) 如果每個假設偏差 20–30%,決策還能成立嗎?逐一測試。
4) 你的安全邊際在哪?情況糟到什麼程度你才會後悔?
5) 在「不順但不極端」的情境下,你能承受嗎?(財務、時間、心理)
6) 你需要怎麼調整這個決策才能讓安全邊際足夠?

提醒:安全邊際不是悲觀,是承認預測必然有誤差。重點是「就算錯了還能撐住」。

互動規則:
1. 一次只問一題,等使用者回答後再進入下一題。
2. 使用者答完所有題目前,不要做總結或下結論。
3. 若答案太抽象、太籠統,請追問一次具體例子或數字後再繼續。
4. 全部答完後,輸出三段:(a) 摘要使用者的關鍵判斷;(b) 你看到的盲點或張力;(c) 一個具體下一步行動建議。
5. 不要替使用者做決定,只把判斷攤開讓他自己決定。

Related Frameworks

FAQ

Does margin of safety apply outside investing?

Yes — and should be. Both Buffett and Munger have stated: "Margin of safety is a core principle of all decision-making." In engineering, it's structural load buffer; in medicine, it's drug dosing room for error; in career, it's not all-in-ing your savings into a startup; in relationships, it's avoiding major commitments at emotional peaks. The core idea: you're not as confident in your forecasts as you think — so design decisions to leave room for "I was wrong," not just to maximize the upside of "I was right."

How big a margin of safety is enough?

Work backward from "cost if wrong" and "the maximum loss you can absorb." Graham's investing rule: "buy at 70% or less of intrinsic value" — meaning even if your valuation is 30% too high, you're still breakeven. The life-decision equivalent question: "if this decision is wrong, what happens?" If the answer is "I'll recover gradually" — small margin is OK. If the answer is "bankruptcy / job loss / divorce / breakdown" — the margin must be large enough to absorb total failure.

How is margin of safety different from "being conservative"?

Conservatism is "lean toward less risk overall" — which can miss good opportunities. Margin of safety is "leave error room in your forecasts" — not reducing action, but making action survivable. Buffett makes huge investments — hardly conservative — but each is built on "even if my valuation is 30% off, I don't lose." In other words: margin of safety isn't avoiding risk — it's making risk non-fatal.

Related studies