Build room for error (margin of safety)
Plan so that being wrong is survivable, not catastrophic.
Why it works
The future has a wider range of outcomes than any plan assumes, and a single ruinous outcome ends the game regardless of average returns. A buffer — savings, slack, conservative assumptions — keeps you in play through the shocks you can’t forecast, which is the precondition for compounding to ever work for you.
How to do it
- Assume your plan is partly wrong and ask: what breaks me if it is?
- Hold a cash/slack buffer sized to let you survive bad luck without being forced to sell or quit.
- Prefer plans you can stick with through a downturn over optimal plans you’ll abandon.
Evidence
Aligns with risk-management and ruin-avoidance principles: avoiding catastrophic, irreversible losses matters more than maximizing expected return when outcomes compound over time. (mechanistic)
This is a well-reasoned principle rather than an outcome study; "how much" room is judgment, not a measured constant.
Sources
- Risk-of-ruin / margin-of-safety reasoning long-standing in finance (Graham; Kelly-criterion intuition on avoiding ruin)
Common mistake
Treating any buffer or unused slack as inefficiency to be optimized away, leaving no cushion for the inevitable surprise.
Practice this with IX Coach
7 days free, then $40/month (~$1.30/day).
More practices for The Psychology of Money, Made Practical
- Treat money as a behavior problem, not a knowledge problem
How you behave under stress beats how much finance you know.
- Define "enough" before you need it
Name the point past which more money no longer buys you anything you value.
- Let compounding do the work (patience)
The biggest results come from time in, not intensity — if you don’t interrupt it.
- Choose reasonable over rational
A plan you can stick with beats an optimal plan you’ll abandon.
- Remember wealth is what you don’t see
Spending signals income; wealth is the money you chose not to spend.
- Save without needing a reason
Saving for "flexibility and options" is reason enough — it doesn’t need a goal attached.