The calm day contract

A friend filled out a risk questionnaire last year. Picked Moderately Aggressive without much hesitation. Said he was fine holding 70 percent in equity, fine riding the swings.

Three weeks into the spring wobble I sat with him while he filled out the same form again. Same job. Same salary. Same goals, same horizon, same emergency fund sitting untouched in the same account. The questionnaire scored him Conservative. Thirty-five percent equity, tops.

Nothing about his actual situation had changed. The only thing that moved was the market, and the number that was supposed to anchor every decision he made moved further than the portfolio it was meant to hold.

He is not unusual. There is a well-known study on exactly this. After the 2008 crash, investors rescored themselves dramatically more risk-averse, and the researchers could not explain it with changes in wealth or income. What they were left with was fear. Risk tolerance, the thing we treat as a stable personal trait, turns out to bend with the last few weeks of the screen.

The expensive part is that the red-day instinct points the wrong way. Selling into a 20 percent fall books the loss and hands you a steeper climb back, because a 20 percent fall needs a 25 percent gain to recover, a 30 needs 43, a 50 needs a full 100. Mistimed exits and re-entries like this have cost investors roughly 15 percent of the return the funds underneath them actually delivered.

So here is what I told him. Fill out the profile on the most boring market day you can find. The calm answers are the honest ones. Then write the allocation down and treat it as a contract, a decision you made while clear-headed that gets to overrule the one you will want to make while scared.

You set your risk profile on the calmest day you can find, precisely so the scariest day does not get to set it for you:

What a scary few weeks looks like in data

The last real test of this was not long ago. Between 31 Dec 2025 and 30 Mar 2026 the Nifty 50 fell from 26,129.60 to 22,331.40, a drop of 14.5 percent, by NSE closing data. Inside that stretch were three of the five worst single days of the past two years: minus 3.26 percent on 19 Mar 2026, minus 2.60 percent on 23 Mar and minus 2.14 percent on 30 Mar.

A questionnaire filled on 30 Mar 2026 would have been filled at the end of a run of the worst days in two years. That is precisely the answer recency bias produces: the last three weeks standing in for the next ten years. A month later the index was 7.5 percent higher. It fell again later in the year, and nobody could have known either move in advance. The point is only that the scared answer described the screen, not the investor.

The climb back is not symmetric

The opening's recovery figures come from one formula. After a fall of d, the gain needed to get back is 1 divided by (1 minus d), minus 1.

Selling at the bottom does not avoid that climb. It hands the climb to cash, which will not make it. The investor who sold at minus 30 and waits for calm usually buys back after a good part of the 42.9 percent has already happened.

How long the climb takes

At an assumed 12 percent a year, a 20 percent fall takes about two years to undo and a 40 percent fall about four and a half. These are the numbers to read on a calm day, so they are not a surprise on a scary one.

Writing the contract

1. Fill the profile on a dull dayNo big move in the past month, no headline you checked twice.
2. Write the split and the worst fallFor example: 70 percent equity, a fall of 25 percent expected at some point.
3. Write the rule for red daysThe SIP continues; rebalance only on the set date or band.
4. Review on life events, not market eventsA new job, a child, a goal date moved. Not a bad month.

The free Learn module on overconfidence, anchoring and recency bias explains why the last few weeks weigh so heavily on judgment, with examples and a quiz. The drawdown recovery calculator gives the climb back for any fall. For the two axis method behind a sound profile, read the lower number is your profile.