Quarnervax · Build useful Scenario Analysis

Scenario analysis sounds straightforward in theory: you imagine a few different futures, assign rough likelihoods to each, and stress-test your thinking accordingly. In practice, most investors quietly build their scenarios around the conclusion they have already reached. The bullish case receives the most detailed treatment, the most generous assumptions, and the most careful sourcing. The bearish case, by contrast, tends to be sketched in quickly, populated with vague risks rather than specific mechanisms, and left without the internal logic that would make it genuinely threatening to the central view. This asymmetry is not usually dishonest — it reflects the natural human tendency to spend more cognitive effort on outcomes we find plausible, and to treat uncomfortable possibilities as boxes to be ticked rather than arguments to be taken seriously. Recognising this habit is the first step towards building scenario work that is actually useful, because a bearish scenario that cannot stand on its own merits will not challenge you when it matters most.

The structural fix is to treat each scenario as a separate investment thesis rather than a variation on a single story. A well-constructed bearish scenario should have its own coherent chain of reasoning: a set of starting conditions, a plausible sequence of events, and a clear account of why the outcome would be worse than the market currently expects. The same applies to the bullish case and to any middle-ground scenarios you construct. When you write each one out in full, rather than simply listing bullet points, you force yourself to identify the load-bearing assumptions — the specific beliefs that, if wrong, would cause the entire scenario to collapse. These are the assumptions worth examining most carefully, because they represent the points of genuine uncertainty in your analysis rather than the details you have already resolved. A useful discipline is to ask, for each scenario, what a well-informed and reasonable person who disagreed with you would say, and then to write that objection into the scenario itself rather than leaving it as an unaddressed afterthought.

Identifying which assumptions matter most is closely related to the question of what new information would cause you to revise your view. This is sometimes called a pre-mortem approach: before committing to a position or a thesis, you imagine that the outcome has already gone badly wrong and ask what would have caused that. The value of this exercise is not that it predicts failure but that it surfaces the specific conditions under which your reasoning breaks down. Once you know those conditions, you can monitor for early signals that they are beginning to materialise, which transforms your scenario analysis from a static document into a living framework for interpreting new information as it arrives. An investor who has thought carefully about the conditions under which their bearish scenario becomes the most likely outcome is far better placed to notice when those conditions are developing than one who treated the bearish case as a formality. This kind of structured vigilance is particularly valuable in environments where information arrives quickly and the temptation to anchor to an existing view is strong.

The final habit worth building is a deliberate review of how your scenarios relate to one another in terms of the assumptions they share. If your bullish and bearish scenarios both rely on the same underlying belief — about the direction of a particular policy, the behaviour of a particular market participant, or the stability of a particular relationship — then you do not actually have two independent scenarios. You have one scenario with a range attached to it, which is a much weaker form of analysis. Genuine scenario diversity requires that at least some of your scenarios challenge assumptions that the others take for granted, because the most significant surprises in investment research tend to come not from the expected variables moving in unexpected directions but from variables that were not considered at all. Building that kind of breadth into your scenario work is difficult and time-consuming, but it is also the quality that separates analysis that genuinely informs decisions from analysis that merely accompanies them.