What it means
Psychologists split the bias into three flavours: overestimation of your own ability, overplacement relative to others, and overprecision, meaning ranges that are far too narrow. The third is the most damaging in finance, because it produces forecasts stated with a confidence the evidence does not support.
The business cost is concrete. Overconfident managers under-provision for risk, set revenue targets that trigger overhiring, and pay too much in acquisitions because they believe they can fix a struggling target faster than anyone before them.
In investing, the classic symptom is excessive trading. An investor convinced of an edge turns the portfolio over repeatedly, paying spreads, commissions and tax each time, and the accumulated friction typically outweighs any advantage the stock picks provided.
The bias is hard to see from the inside because good outcomes feel like proof of skill while bad ones feel like bad luck. That asymmetry means experience alone rarely cures it, and confident people are often rewarded socially for the very certainty that makes their estimates worse.
Practical fixes work on the process rather than the person. Requiring ranges instead of point estimates, keeping a written forecast log, running a pre-mortem where the team assumes the project has already failed, and building base rates from past projects all pull estimates back towards reality.
In practice
Real-world examples.
Example
A founder projects 40% revenue growth for three consecutive years based on one strong quarter, then hires 18 people against it. Growth comes in at 12%, and the company is forced into redundancies nine months later.
Example
A construction firm bids for a hospital fit-out at a price that assumes no weather delays and no design changes, because the last two projects ran smoothly. The job overruns by 14 weeks and the fixed-price contract turns a projected profit into a loss.
Example
A private investor reviews her broker statement and finds 63 trades in a year against a buy-and-hold benchmark she has underperformed. The stock selection was slightly positive, but costs and tax turned it negative.
Think of it
“Overconfidence is thinking you know more than you do-excessive faith in your own abilities.
Formula
Calculation
There is no single formula, but overconfidence is measured by calibration: the gap between stated confidence and actual hit rate, plus the trading cost it causes.
An investment committee makes 20 forecasts over a year, each stated with 90% confidence. Only 12 turn out correct, so the hit rate is 12 / 20 = 0.60, or 60%, against a stated 90%. The calibration gap is 90% - 60% = 30 percentage points, a clear signal of overprecision rather than bad luck on one call.
The same committee's confidence drives turnover in a $400,000 portfolio: it trades the whole portfolio one and a half times a year, and each round trip costs about 0.5% in spread and commission. Annual cost drag is 150% x 0.5% = 0.75%, which is $400,000 x 0.0075 = $3,000 a year. Over ten years, at that rate on the same balance, the committee's certainty costs roughly $30,000 before any effect on the returns themselves.Case study
Seen in the real world.
This is an illustrative, entirely fictional example. Marram Analytics, an invented data firm, won a $2,400,000 platform contract after its delivery lead promised a nine-month build, having quietly halved his team's own estimate because he "knew the codebase". The board approved the bid without asking for a range or a downside case.
The project took sixteen months in this fictional scenario, consuming $3,100,000 of internal cost and triggering $180,000 of late-delivery penalties. A post-project review found the estimate had been made in a single afternoon with no reference to the firm's own history, where the average project had run 45% over its first estimate.
Marram's illustrative fix was procedural rather than personal: every bid above $500,000 now requires a range, a written base rate drawn from past projects, and a pre-mortem session. The next three bids were priced roughly 30% higher, the firm lost one of them, and delivered the other two at a profit.
Watch out
Common mistakes.
- Treating overconfidence as a personality flaw in other people rather than a predictable pattern that affects almost every forecaster.
- Asking for a single number in a plan instead of a range, which hides how uncertain the estimate really is.
- Reading a run of good results as proof of skill, when a rising market or a favourable run of luck explains most of it.
Questions
People also ask.
How do I tell overconfidence from justified confidence?
Keep a forecast log and compare stated confidence with actual outcomes over at least twenty calls, since only the track record can settle it.
Does overconfidence always destroy value?
No, it helps people start ventures and persist through setbacks, but it should be kept away from pricing, provisioning and capital allocation decisions.
What is the quickest practical remedy?
Run a pre-mortem: assume the plan has already failed and ask the team to explain why, which surfaces risks that optimism suppressed.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%