“Price target: €187.” Analyst forecasts sound that precise, and they miss just as regularly. That’s no criticism of analysts; it’s simply the nature of the thing. A company’s future isn’t a number but a range of possible paths. Investing always means deciding under uncertainty. The good news: you can calculate with uncertainty. That’s exactly what scenario thinking is for: bull case, base case, bear case. In this guide, you’ll build this tool yourself, step by step.
Why point forecasts fail
A point forecast (“the fair value is €187”) pretends the future can be condensed into a single number. Yet a stock’s value depends on assumptions that are themselves uncertain: revenue growth, margins, interest rates, competition. Even small deviations compound. An example: if a company grows at 8% instead of 10% a year for ten years, revenue ends up roughly 17% lower. And because valuation models like the DCF method discount future earnings, a deviation like that hits fair value disproportionately hard.
Then there’s psychology: a single number creates false precision. It feels like knowledge, even though it describes only one of many possible futures. If you cling to a point forecast, every deviation catches you off guard. If you think in scenarios, you’ve already played through the deviation in advance.
Bull, base, bear: how to build three scenarios
A scenario isn’t a fantasy story but a consistent combination of drivers, assumptions and the value that follows from them. Three are almost always enough: an optimistic path (bull), a realistic one (base) and a pessimistic one (bear).
Step 1: Find the two or three real drivers
Every investment thesis hinges on a few key levers. For a software company, that might be user growth and pricing power; for an auto supplier, order intake and margin; for a bank, interest rates and loan defaults. Ask yourself: which two or three variables will decide, five years from now, whether this investment worked? Everything else is noise. Our guide Learn stock analysis shows how to derive these drivers systematically from the business model and key metrics.
Step 2: Set concrete assumptions for each scenario
Now each scenario gets numbers. Important: the bull case isn’t “everything goes perfectly” but a plausibly good path, such as the upper end of what the company has achieved before or what the best competitors demonstrate. Conversely, the bear case isn’t the end of the world but a plausibly bad path: growth halves, the margin falls to the industry average, a major customer leaves. You can be aware of extreme scenarios (total loss, a tenfold gain), but they don’t belong at the center of the calculation.
Step 3: Estimate rough probabilities
Nobody knows exact probabilities, and you don’t need to. A rough, honest weighting in steps of ten is enough, for example 25% bull, 50% base, 25% bear. What matters is less the exact number than the debate with yourself: if you want to give the bull case 60%, you need evidence for it, not just enthusiasm. Rule of thumb: the base scenario should be the most likely one; if it isn’t, it’s not your base scenario.
Want to see this applied to a real stock? The AI analysis calculates exactly these metrics for more than 1,900 stocks.
Analyze a stockExpected-value thinking: a worked example
The expected value combines the three scenarios into a basis for decisions: each scenario value is multiplied by its probability, and the results are added up. Suppose a stock costs €100 today and you estimate the following over a three-year horizon:
| Scenario | Assumption | Value per share | Probability | Contribution |
|---|---|---|---|---|
| Bull | Growth holds, margin rises | €160 | 25% | €40.00 |
| Base | Solid growth, stable margin | €110 | 50% | €55.00 |
| Bear | Growth collapses, margin falls | €60 | 25% | €15.00 |
Expected value: 40 + 55 + 15 = €110, about 10% above today’s price. That’s positive, but not a sure thing: in the bear case, you face a 40% loss. Whether a 10% expected return is worth that risk to you is a second, separate decision, and a question of position size, in other words of portfolio diversification.
The real value of the calculation shows when you vary the assumptions. If the bear case is €40 instead of €60 (say, because of high debt), the expected value drops to €105, and the margin of safety is almost gone. If, on the other hand, the stock trades at €85 instead of €100, the same estimate delivers an expected return of nearly 30%. That turns “I like this stock” into a testable statement: At what price is this stock a good deal, and at what price is it no longer one?
A second important insight: two stocks with the same expected value are not equally good. A +10% expected value with a range of −10% to +25% is something completely different from +10% with a range of −60% to +90%. Asymmetry matters: you’re looking for situations where the bear case is limited and the bull case is large, not the other way around.
Scenarios as protection against confirmation bias
Confirmation bias, the tendency to collect only evidence that supports your own view, is one of the most expensive investor mistakes of all. Scenario thinking counters it almost mechanically: if you have to write out a bear case, you’re forced to research the strongest counterarguments yourself instead of brushing them aside. “The bears don’t understand the business model” becomes a concrete question: what exactly would have to be true about their arguments, and what would it cost?
Practical tip: write the bear case so convincingly that a skeptic would sign off on it. If you can’t come up with anything substantial, you’ve either found an exceptional company or, far more likely, not looked hard enough yet.
Pre-mortem: what would have to happen for the thesis to break?
A pre-mortem flips the perspective: imagine it’s three years later and the investment has failed; the stock is 50% lower. What happened? In practice, this backward-looking question produces more honest answers than the forward-looking “What could go wrong?”, because it treats failure as a fact rather than a distant possibility.
Write down two to four concrete breaking points for your thesis, for example: “The largest customer (30% of revenue) doesn’t renew,” “gross margin falls below 40% for two quarters in a row,” “a competitor undercuts prices by more than 20%.” The key: these breaking points are observable. The pre-mortem thus turns into checkpoints you tick off with every quarterly report. If a breaking point occurs, you sell or recalculate, without drama, because you made the decision while you were still calm.
The market prices in scenarios; the question is which ones
A stock price is itself something like an expected value: the weighted sum of what all market participants consider possible. That’s why the right question is never “Is this a good company?” but “Which scenario is already in the price?” If a quality stock trades at 45 times earnings, the market has largely anticipated the bull case: even if it materializes, little return is left, and the base case is already enough for disappointment. Conversely, a stock at 6 times earnings may already have a very dark scenario priced in; in that case, you may profit simply because things go badly rather than catastrophically.
That’s exactly why structured analyses work with ranges instead of point targets. The AI report from aktienanalyse.ai also provides a fair value range with an explicit bull and bear case, so you can apply your own probabilities to it instead of adopting a single number.
Conclusion: calculate instead of guess
Point forecasts fail because they hide uncertainty. Scenarios make it visible and manageable: identify two or three real drivers, set plausible assumptions for each scenario, weight them roughly, compare the expected value with the current price, and use a pre-mortem to record how you’ll know you’re wrong. That takes about an hour per stock and changes the quality of your decisions more than any additional stock-price app. Not because the numbers are exact, but because you’re forced to state your assumptions before the market tests them.