Lesson 2 of 9

Loss aversion, anchoring and herding

Step 3 of 6

Worked example

"I'll sell when it gets back to what I paid." Take a paper position in Nvidia bought at the close on 21 Oct 2026, $221.97. On 19 Nov 2026 it closed at $145.98 (Yahoo, closing prices; Module 16 lesson 07).

Loss so far      = 145.98 ÷ 221.97 − 1 = −34.2%
Gain to get back = 221.97 ÷ 145.98 − 1 = +52.1%

How the biases could bite. Loss aversion makes selling at −34% feel like locking in pain. The disposition effect turns $221.97 into the exit line: "I'll sell at break-even." That quietly sets a +52% target before you'll act, chosen by the date you happened to buy, not by anything about the business.

Why the cost basis doesn't belong in the decision. The market doesn't know what you paid. Someone who bought at $221.97 and someone who bought at $100 own the same share at $145.98, with the same future. The useful question is the same for both:

Knowing what I know today, would I buy this at $145.98, and at this size?

If yes, the sell card from Module 16 lesson 07 decides what to do. If no, "waiting to get even" is just holding a position you wouldn't buy. (Selling can trigger taxes, which reduce returns; check your local rules. That's a real cost, but a separate line, not a reason to anchor.)

And herding. On 21 Oct the buy may have felt urgent because everyone seemed to own it. Neither the crowd's excitement then nor its fear now is information about the business.

This shows how biases could distort the decision, not whether holding or selling Nvidia is right.