Actionable insights — How To Mentally Handle Tough Times

1. Diagnose a bubble by the factor spread, not the index level

1. Stop asking whether the index is expensive — that question has no reliable answer and no timing content.
2. Measure the spread between momentum/high-beta and quality instead. Extreme, sustained divergence between the two is the observable signature.
3. Compare the current spread with historical episodes to see whether it is unusual rather than merely uncomfortable.
4. Use the reading to set expectations and position sizing, not to sell — the diagnosis says where you are, not when it ends.

• The self-serving fit. A quality manager diagnosing "quality underperformance means bubble" is describing their own drawdown as evidence — the reasoning is sound but the incentive is obvious.

2. Benchmark your drawdown against named peers before diagnosing yourself

1. List the managers running the closest approach to yours and find their year-to-date numbers.
2. Compare your own drawdown with theirs. Similar magnitude across independent, skilled practitioners means the factor is being repriced, not that you erred.
3. Note what those managers are doing in response, which is more informative than what they are saying.
4. Re-run the comparison periodically rather than only when it is comforting.

• Comfort substituting for analysis. Shared pain explains the drawdown; it does not establish that the strategy will recover.

3. Watch where retail money moves during a drawdown — and go the other way

1. Track the direction of flows between individual stocks and index funds during and after declines.
2. Note the timing: the switch into index funds tends to happen when individual names are at their cheapest, and the switch back after the recovery.
3. Use it as a contrarian signal on your own behaviour first — the pressure you feel to make that switch is the signal.
4. Decide the allocation between individual names and funds in advance, so a drawdown cannot renegotiate it.

• Who is telling you. The publication making this argument also runs an ETF portfolio, so it is describing a switch between two of its own products.

4. Remove the portfolio value from view — in good times as well as bad

1. Use the broker's own settings to hide the portfolio value and the daily profit and loss on login.
2. Apply it permanently, not situationally: hiding the number only when it is falling is an emotional response, not a system.
3. Replace the hidden number with a slower one you check on a fixed schedule — Owner's Earnings or free cash flow.
4. Test the design against the property analogy: you own a house for decades without a daily valuation, and that is why you keep it.

• Hiding the number instead of monitoring the businesses. The point is to substitute a better metric, not to stop looking altogether.

5. Choose a strategy on two axes — fits you, and has worked

1. Require both conditions: the approach must suit your temperament, and it must have a long-run record of outperformance. Either alone fails.
2. Test the first condition honestly by asking whether you could hold the positions through a multi-year drawdown without external confirmation.
3. Remember that borrowed ideas do not come with borrowed conviction — "you can copy someone's stock idea, but you can never copy someone's conviction."
4. Do not switch strategies during a drawdown; the switch is being made at the point of maximum discomfort, which is the worst available moment.

• "Suits me" being defined after the fact. The test is whether you can hold, not whether the story is appealing.

6. Split the world into what you control and what you do not — then work only on the first

1. Accept the split: sentiment, multiples and timing are outside your control; what you buy, who runs it, and what you pay are inside it.
2. Write the three controllables as purchase criteria — wonderful business, managers with skin in the game, fair valuation.
3. Track the three measurement dials on a schedule: portfolio fundamentals versus the index, Owner's Earnings, free cash flow.
4. Judge yourself on the controllables, since "in the short term, the market is a voting machine. In the long term, it's a weighing machine."

• The dials being shown without numbers. A chart with no scale cannot be checked against the previous month's claim.

7. Zoom out — with the honest version of mean reversion

1. When a period feels unprecedented, find the historical analogue and read what actually happened afterwards.
2. Apply reversion to the mean symmetrically: extreme good performance reverts as surely as extreme bad performance.
3. Extend the chart until the current episode becomes small — that is the operational meaning of "zoom out."
4. Keep the recovery time realistic rather than assuming it is quick.

• "Always" doing too much work. Mean reversion says the level recovers eventually; it says nothing about whether you are still invested when it does.