A Base Case Is Not a Risk Analysis

The expected outcome tells leaders where the centre of the forecast sits; it does not reveal how far reality may move away from it.

Investment proposals usually contain a base case.

Revenue, cost, demand, delivery timing, residual value and other assumptions are converted into one expected financial result. That result may be a net present value, internal rate of return, cost estimate or benefit forecast.

The base case is useful.

It is not risk analysis.

The supplied Hargitay material defines investment risk around the possibility that actual outcomes diverge from expected outcomes. It then introduces several complementary methods for understanding that divergence: expected value and variance, sensitivity analysis, scenarios, simulation and market-related measures.

The Week 4 study material reinforces the same point and explicitly notes that risk-analysis methods are rarely sufficient when used alone.

The executive implication is simple:

A decision under uncertainty requires information about the distribution of outcomes, not only the most convenient point estimate.

The Strategic Context

A single forecast compresses uncertainty.

Suppose a project has an expected NPV of $50 million.

That number could describe several very different risk profiles.

One project may have plausible outcomes tightly clustered between $40 million and $60 million.

Another may have a large probability of losing $100 million and a smaller probability of producing a very large upside, with the same expected value.

Those are not equivalent investments.

The expected value alone cannot tell leadership:

• probability of loss;
• probability of meeting the approved business case;
• potential capital overrun;
• severity of downside;
• dependence among variables;
• or the chance that several adverse conditions occur together.

A base case therefore answers one question:

What outcome are we using as the central planning assumption?

Risk analysis answers another:

What range of outcomes should leadership be prepared to absorb?

What Leaders Commonly Misread

The first error is to describe a conservative base case as risk analysis.

A conservative estimate may reduce optimism bias, but it does not show the consequences of uncertainty.

The second error is to run a few sensitivity tests and conclude that the risk has been quantified.

Hargitay describes sensitivity analysis primarily as a way to identify which factors can materially change investment outcomes. It is valuable because it answers “what if?” questions and can identify break-even values.

But one-variable-at-a-time sensitivity analysis does not automatically reveal the probability of those changes or the interactions among variables.

The third error is to treat optimistic, realistic and pessimistic scenarios as a complete probability model.

Scenarios are useful because they combine variables into coherent states of the world. Their weakness is that probabilities may be absent, crude or subjective.

The fourth error is to assume that simulation eliminates judgement.

Simulation is only as credible as the model, relationships and probability distributions entered into it.

A sophisticated simulation built from weak assumptions can produce highly precise nonsense.

Reframing the Issue

Risk analysis should be treated as a progressive increase in visibility.

Each analytical technique answers a different question.

The aim is not to use every method on every project.

The aim is to match analytical depth to decision consequence.

Sensitivity Analysis Identifies Leverage Points

Sensitivity analysis is valuable because it is easy to understand.

Change one important input and observe what happens to the result.

For example:

• What if demand is 15 per cent lower?
• What if construction cost is 20 per cent higher?
• What if commissioning is delayed by six months?
• What if the commodity price falls?
• What if adoption reaches only half the forecast?

This identifies decision leverage.

If a small change in one assumption destroys the investment case, leadership has learned something important.

But sensitivity does not tell leaders whether that change is likely.

That is why sensitivity is the beginning of risk analysis, not the end.

Related article: Sensitivity Analysis Should Change the Decision, Not Decorate the Appendix

Scenarios Reveal Coherent Futures

Scenario analysis groups assumptions that plausibly move together.

An economic downturn may combine:

• lower demand;
• weaker pricing;
• tighter credit;
• delayed customer payments;
• and supplier stress.

A technology acceleration scenario may combine:

• faster adoption;
• higher implementation demand;
• shorter asset life;
• and stronger competitor response.

The Hargitay and Week 4 materials use optimistic, realistic and pessimistic combinations as simplified examples.

The strategic value is not in those labels themselves.

It is in forcing decision-makers to consider coherent alternative futures rather than changing variables in isolation.

A useful scenario should therefore explain why its assumptions belong together.

Probability Changes the Decision Conversation

McKinsey's supplied capital-project article provides a practical illustration of the value of probability distributions.

Rather than presenting only a baseline NPV, its example distinguishes expected value, probability of breaking even, probability of meeting the original baseline and downside exposure.

That changes the nature of the executive conversation.

A project may have attractive expected economics while only having a modest chance of achieving the original business case.

Leadership then has more useful questions to ask:

• Is the downside financeable?
• Can we mitigate the largest driver?
• Should the project be staged?
• Can contracts transfer or cap some exposure?
• Is there an option to exit?
• What additional evidence would most improve the decision?

The analysis becomes a tool for designing the investment, not merely approving it.

Correlation Matters

Hargitay also highlights correlation among period cash flows.

The broader lesson applies well beyond investment mathematics.

Uncertain variables often move together.

A market downturn may reduce volume and price simultaneously.

A project delay may increase labour cost while postponing benefits.

A regulatory change may create both redesign cost and schedule delay.

If risk analysis treats these effects as independent when they are not, downside exposure may be understated.

Conversely, assuming everything deteriorates together can make the model unnecessarily pessimistic.

The relationship among variables matters as much as the range of each variable.

Simulation Is Powerful but Conditional

The Week 4 material describes Monte Carlo simulation as repeatedly solving an evaluation model using values selected from probability distributions, producing an overall distribution of NPV outcomes.

This can provide richer information than a handful of scenarios.

But the source also makes an important caution: reliability depends on how closely the model represents reality.

Simulation should therefore not be used to create decorative complexity.

It is useful when:

• several variables are genuinely uncertain;
• interactions matter;
• the decision is consequential;
• data or defensible expert judgements support distributions;
• and leadership can interpret the resulting outputs.

It is less useful when assumptions remain fundamentally unknown and the model simply assigns artificial probabilities.

Decision Framework

Choose risk-analysis depth according to the decision.

Low consequence or reversible

Use:

• base case;
• key sensitivities;
• break-even thresholds.

Material but manageable

Add:

• coherent scenarios;
• downside funding requirements;
• explicit assumptions and triggers.

Large, irreversible or strategically consequential

Consider:

• probability distributions;
• simulation;
• correlation;
• capital-at-risk;
• and portfolio interaction.

Across all levels, require leadership to understand:

1. Which assumptions drive the result?
2. What is the plausible downside?
3. What is the probability of failing to meet the business case?
4. What additional capital might be required?
5. Which risks can be mitigated before commitment?
6. Which risks must simply be accepted?

From Strategy to Execution

Immediate action

Stop presenting the base case as the whole financial truth.

For major proposals, show at least one downside case and identify the variables that most influence value.

Separate probability of breakeven from probability of achieving the approved plan.

Medium-term capability building

Standardise risk-analysis methods across the investment portfolio.

Build internal capability in scenario design, probability assessment and simulation where justified.

Record which assumptions proved wrong after delivery and feed that evidence back into future models.

Long-term strategic positioning

Create a culture in which uncertainty is visible without becoming paralysing.

Decision-makers should be able to approve an investment while acknowledging that the base case may not occur.

The quality of governance lies not in eliminating uncertainty, but in understanding what outcomes the enterprise is prepared to absorb.

Signals to Monitor

Risk analysis is weak when:

• every proposal produces one precise NPV with no distribution;
• sensitivity ranges are chosen arbitrarily;
• project teams can explain the base case but not the downside;
• scenario probabilities are presented as facts without evidence;
• simulation outputs are accepted because they look sophisticated;
• correlated risks are treated as independent;
• or a project is described as low risk simply because its expected value is high.

Questions for the Leadership Team

1. What is the probability that this project fails to meet the approved base case?
2. Which variable has the greatest leverage over the result?
3. What combination of adverse events creates the most credible severe downside?
4. How much additional capital would we need under that scenario?
5. Which assumptions are supported by evidence and which depend mainly on judgement?
6. What would we change about the project if the downside were treated as real rather than theoretical?

Closing Perspective

A base case is necessary for planning.

It is dangerous when mistaken for certainty.

Investment risk is the possibility that reality diverges from the expected outcome. Decision-grade analysis therefore needs to reveal not only the centre of the forecast, but the shape, drivers and consequences of that divergence.

The objective is not to replace judgement with probability.

It is to give judgement a more truthful picture of what the organisation is actually choosing to risk.