Cognitive Biases

Cognitive biases: systematic errors in thinking
The human brain uses heuristics—mental shortcuts—for quick decision-making. These shortcuts often lead to systematic errors known as cognitive biases.

Key cognitive biases of investors

Bias	Description	Example in investing
Overconfidence	Overestimating one's abilities and knowledge	"I can beat the market"
Anchoring	Tying decisions to the first piece of information	"The stock was $100, now $50—it's cheap!"
Confirmation Bias	Seeking information that confirms one's opinion	I only read bullish reports
Recency Bias	Overweighing recent events	"The market has risen for 3 years, so it will keep rising"
Hindsight Bias	"I knew it all along"	The 2008 crisis was obvious (not really)
Availability Heuristic	Easily recalled = probable	After a crash, I'm afraid to fly (although statistically safe)

Overconfidence: in detail

Research shows:

80% of drivers consider themselves better than average (which is mathematically impossible)
70% of active managers underperform the index
Traders with high activity show worse results
Terrance Odean (1999): The more trades— the worse the results.
Overconfidence → over-trading → costs → underperformance.

Anchoring: examples

Situation	Anchor	Error
Stock fell from $100 to $50	$100 = "correct" price	$50 may be the fair value
Analyst forecasts $80	$80 becomes the anchor	Own analysis is ignored
IPO price $25	$25 = value	IPO price not connected to value

Confirmation Bias: mechanism

Form an opinion ("the stock will rise")
Search for confirming information
Ignore contradicting information
Become more confident in the opinion
Are surprised when wrong

Debiasing Techniques

Bias	Defense technique
Overconfidence	Track record, post-mortems, probability estimates
Anchoring	Start from scratch, ignore prior price
Confirmation Bias	Actively seek disconfirming evidence
Recency Bias	Long historical data, mean reversion
Hindsight Bias	Document predictions before events

Investment Checklists

Checklist before each investment:

What is my thesis? Write down before purchase
What can go wrong? (Pre-mortem)
Under what conditions will I sell?
What information would change my opinion?
Did I look for arguments against?
Am I too confident?

Red Team: institutional solution

Assign a "devil's advocate" on every investment committee:

Task: find reasons NOT to invest
Rotate the role among members
Make it a formal part of the process

Probabilistic Thinking

Replace binary ("will/won't happen") with probabilities:

Binary	Probabilistic
"The stock will rise"	"70% probability of 15%+ growth"
"There will be no recession"	"30% chance of recession in 12 months"

CIO Recommendations

Document—rationale before the deal
Look for counterarguments—actively
Use checklists—every time
Track record—honest accounting of results
Humility—the market is smarter than you