The Sleep-at-Night Test
The Sleep-at-Night Test
If you can't sleep through a 30% crash without checking your portfolio daily or panicking, that allocation is wrong. No questionnaire changes this fact.
Key takeaways
- The sleep-at-night test is empirical: can you actually sleep during a crash, or does market volatility keep you up?
- A portfolio that costs you sleep is, by definition, too aggressive for your willingness, regardless of what your time horizon allows.
- This test is personal and honest, unlike questionnaires which measure aspirations.
- Most investors who claim 80/20 tolerance actually need 60/40 or lower to sleep.
- The optimal allocation is the highest equity percentage you can hold without losing sleep, not the highest allowed by time horizon.
Why sleep matters
Sleep deprivation degrades decision-making. An investor who's anxious and sleep-deprived during a crash is more likely to:
- Check the portfolio obsessively (emotionally draining)
- Second-guess the allocation (undermining the plan)
- Sell at the bottom (locking in losses)
- Miss the recovery (and regret it for decades)
Conversely, an investor who sleeps soundly through volatility can:
- Stay the course without obsessive checking
- Rebalance or buy during crashes (buying low)
- Avoid panic-selling
- Realize full market recovery
Sleep quality is not a proxy for risk tolerance; it's a measure of whether the allocation is actually suitable for your psychology.
How to run the test
Step 1: Identify the allocation you're considering. Say you're looking at 80/20 stocks/bonds, or 70/30, or whatever matches your time horizon.
Step 2: Calculate the 30% drawdown scenario. In a typical bear market, the global stock market falls 20–35%. Calculate what your portfolio would be worth after a 30% decline:
- £200,000 portfolio, 80/20 allocation = £160,000 in equities, £40,000 in bonds.
- 30% crash in equities = £160,000 → £112,000 (loss of £48,000).
- Total portfolio: £112,000 + £40,000 = £152,000 (24% total loss).
Or use the simpler shortcut: a 80/20 portfolio loses roughly 24% in a 30% stock crash. A 70/30 portfolio loses roughly 21%. A 60/40 loses roughly 18%.
Step 3: Imagine this loss, right now, today. Not in theory. Today. Your £200,000 is now £152,000. Permanently? You don't know yet. The news says recession, unemployment rising, more losses expected.
Can you sleep tonight?
Step 4: Be honest about what "sleep" means.
- Sleeping soundly, checking portfolio weekly at most = you can handle this allocation.
- Lying awake, checking daily, anxious when markets close = allocation is too aggressive.
- Wanting to sell to "cut losses" = allocation is far too aggressive.
- Convinced the losses are temporary and staying the course = you can handle this allocation.
Real examples: the sleep-at-night test in practice
Example 1: The anxious banker. A 45-year-old with £600,000, income of £80,000, and a 20-year horizon. Time horizon allows 70/30 or even 80/20. But when asked, "If this portfolio fell from £600,000 to £480,000 (20% loss), would you sleep?" the answer is "No, I'd be checking it hourly and probably want to sell."
Allocation: 50/50, even though time horizon allows more. The sleep test overrides the questionnaire. A 50/50 portfolio falling 15% to £510,000 is still stressful, but it doesn't cost sleep.
Example 2: The stoic engineer. A 52-year-old with £400,000, stable income of £60,000, and a 15-year horizon (to early retirement at 67). Time horizon suggests 70/30. When asked about a £400,000 → £280,000 crash (30% loss), the response is: "I'd be annoyed, but it wouldn't stop me sleeping. I know I have 15 years and crashes recover. I might even see it as a chance to buy."
Allocation: 80/20 or even higher. The willingness backs the time horizon. The sleep test is clean.
Example 3: The retiree with mixed needs. A 70-year-old with £800,000, pension of £25,000, and annual spending of £45,000 from the portfolio (4% initial withdrawal rate). Time horizon is mixed (immediate income needs, legacy wealth for 30+ years).
Immediate needs (next 5 years, £225,000) should be in bonds/cash—0% stocks. Legacy wealth (20+ years out) can be 80/20. Overall allocation: 40/60 or 50/50 depending on how the portfolio is segmented.
A crash to £560,000 is painful (retirement lifestyle threatened?), but if the portfolio is properly bucketed, the near-term cash is untouched, and only the legacy portion drops. Sleep is possible.
The sleep test is superior to questionnaires
A questionnaire asks: "How would you react to a 30% loss?" A sleep test asks: "Could you actually live with a 30% loss?"
The sleep test is superior because:
- It's empirical: Either you sleep or you don't. No self-deception.
- It's personal: Unaffected by social desirability or aspirational answering.
- It's actionable: If the test fails, you lower the allocation. Simple.
- It's predictive: Investors who can sleep through crashes do stay the course. Investors who can't do panic-sell.
What if you fail the sleep test?
If a 70/30 portfolio would cost you sleep, don't hold 70/30. Simple.
Options:
- Lower the equity allocation to 60/40, 50/50, or lower. Test again. Where's the boundary where you can sleep?
- Segment the portfolio. Keep the long-term portion aggressive (80/20), but hold the near-term portion conservative (0% stocks). This way, crashes don't threaten immediate spending, and sleep is easier.
- Adjust the timeline. If you need to retire in 10 years and your sleep test says you can't hold 70/30, can you extend to 12–15 years? Extra years allow lower equity allocation while still meeting goals.
- Reduce spending goals. Lower spending need = lower required return = lower equity allocation = easier to sleep.
The point is: the allocation that costs you sleep is wrong, even if it's justified by time horizon. No formula overrides sleep quality.
The sleep test in context
The sleep test is not the only factor—capacity and need still matter. But when willingness is the limiting factor, it's the most honest way to measure it.
A 30-year-old with 35 years to retirement might capacitively support 95/5, but if 95/5 costs them sleep (perhaps due to a family history of loss or anxiety), then 70/30 is the right allocation, even if it's technically conservative for their age.
Conversely, a 65-year-old in early retirement might emotionally be comfortable with 80/20, but if capacity is insufficient (low assets, high spending need), the allocation must be lower regardless of sleep-at-night comfort.
The formula: Find the allocation that is supported by capacity, required by need, and tolerable by willingness. That's the right allocation.
If one factor blocks the others, the plan needs adjustment before you invest.
Testing specific allocations
Here's how to run the test for different allocations:
| Allocation | 30% Stock Crash | Portfolio Loss | Can You Sleep? |
|---|---|---|---|
| 20/80 (conservative) | 6% portfolio loss | £200,000 → £188,000 | Likely yes |
| 40/60 (moderate) | 12% portfolio loss | £200,000 → £176,000 | Usually yes |
| 60/40 (balanced) | 18% portfolio loss | £200,000 → £164,000 | Often uncomfortable |
| 80/20 (aggressive) | 24% portfolio loss | £200,000 → £152,000 | Rarely yes |
| 100/0 (very aggressive) | 30% portfolio loss | £200,000 → £140,000 | Almost never yes |
Most investors can sleep through a 12% loss (40/60). Many can tolerate 18% (60/40). Few can tolerate 24% (80/20) unless they've explicitly lived through a crash and know they can.
The sleep-at-night checklist
Before finalizing an allocation:
□ I can imagine a 30% stock market crash. □ I can calculate what my portfolio would be worth (use the table above). □ I have pictured this number clearly. □ I believe I could sleep that night without checking the market. □ I believe I could stay the course for the next week of bad news. □ I believe I could rebalance (buy more equities) at the bottom, not sell. □ I've lived through at least one similar crash before and actually held.
If any box is unchecked (especially the last one), lower the allocation one step and re-test.
The flowchart
Related concepts
Next
The sleep-at-night test is simple, honest, and predictive. It tells you whether your allocation is psychologically sustainable. But there's a distinction you might have missed: some portfolio declines are temporary volatility (which recovers), and some are permanent losses (which don't). That difference matters enormously—and it's often misunderstood.