What it means
Some probabilities come from data. If a factory has made a million parts and 5,000 were faulty, the chance of a fault is 0.5% and nobody needs to guess.
Many business questions have no such record. What is the chance a new product launches successfully, a rival cuts prices next quarter or a regulator approves a licence?
Here you have to estimate, and the estimate reflects what you know, what you have seen before and how you read the situation. Because the number comes from a person, two people can reasonably give different answers and neither is automatically wrong.
The estimate should be updated when new evidence arrives, and a good forecaster revises it as the facts change instead of defending the first guess. To use it well, you write the number down, state the reasoning and test it against what actually happens.
Teams often ask several people to give their own figures independently and then compare them, which exposes optimism, overconfidence and wishful thinking. Reviewing past estimates against outcomes is the way to get better at it.
Psychology plays a part too. People tend to be overconfident about their own projects and to underweight outcomes that are unpleasant to think about, so a probability given by the person who owns the plan is often too high.
Asking someone outside the project to give a number, or asking how often similar projects have succeeded elsewhere, helps to correct for that. Subjective probabilities feed into expected value calculations, scenario planning and decision trees.
They turn a vague feeling such as "this is probably fine" into a number you can multiply and compare, while still being honest that the number is a judgement rather than a measurement.
In practice
Real-world examples.
Example
A founder is asked by investors how likely it is that her start-up wins a major retail contract. She says 40% based on three similar bids she has run, and the investors use that figure in their model. They also ask her to explain which factors would raise or lower it, so the number can be updated as the bid progresses.
Example
An insurance underwriter reviews an unusual risk, such as cover for an event with no claims history. She sets her own probability of a claim after talking to engineers and checking similar risks, and prices the policy accordingly.
Example
A construction firm estimates a 30% chance that bad weather delays a project by more than a month. The project manager uses that number to decide whether to build in a contingency of $90,000. She reviews it each month as the weather forecasts and site progress change.
Formula
Calculation
Subjective probabilities are usually used to calculate an expected value:
Expected value = Sum of (Probability x Outcome)
A manager assigns a 25% chance that a new service earns a $1,000,000 profit, a 50% chance it earns $400,000 and a 25% chance it loses $200,000. The expected value is (0.25 x $1,000,000) + (0.50 x $400,000) + (0.25 x -$200,000) = $250,000 + $200,000 - $50,000 = $400,000. The probabilities add to 100%, as they must, and the result is a weighted average rather than a prediction of what will actually happen.Case study
Seen in the real world.
Brightfield Foods is an illustrative, fictional manufacturer weighing whether to open a second plant. The finance team had no history of opening a plant in a new region, so it asked five senior managers to give their own probability that the plant would reach break-even within two years.
The answers ranged from 30% to 70%, and the spread itself was informative: the most optimistic managers had not allowed for a delay in getting permits. After a discussion, the group agreed on a figure of 50% and built a model on that basis.
In this illustrative story, the plant took three years instead of two to break even, but the team had set aside enough cash for the slower case. The probability was not proved right or wrong by a single outcome, yet the exercise prompted a plan for the downside.
Watch out
Common mistakes.
- Treating a subjective probability as a measured fact rather than an informed judgement.
- Giving 0% or 100% to an event that is merely unlikely or very likely, which leaves no room for being wrong.
- Never revisiting the estimate after new information arrives.
Questions
People also ask.
Is a subjective probability just a guess?
It is an estimate based on evidence and experience, and it improves when it is reasoned, written down and checked against outcomes.
How is it different from objective probability?
Objective probability comes from data or known physical chances, such as the odds on a fair coin, while subjective probability reflects belief.
Can subjective probabilities be combined?
Yes, many teams average several independent estimates, which tends to reduce the effect of one person's bias.
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%Related