Forecasting 101
A Beginner’s Guide to Superforecasting Techniques
Forecasting may sound intimidating, but with the right tools, anyone can improve their ability to predict the future. This guide is based on the principles of Superforecasters, individuals who consistently outperform others in making accurate predictions. Let’s dive into the techniques to help you forecast like a pro.
1. The Fermi-Size Approach: Breaking It Down
Named after physicist Enrico Fermi, this method involves breaking a big, complicated question into smaller, more manageable pieces. Think of it like solving a puzzle step by step. Steps: Start with a big question: For example, “How many piano tuners are there in New York City?” Break it down: Ask simple sub-questions like: How many people live in New York City? How many households own pianos? How often do pianos need tuning? Estimate each piece: Use approximate numbers that feel reasonable. Multiply the pieces back together: Combine your answers to get a total estimate. Tip: The Socratic Method (asking yourself and others pointed questions) helps refine your thinking as you break problems down.
2. The Outside View: Start Broad
The outside view is about grounding your predictions in reality by using historical data or base rates. It’s your anchor to avoid wild guesses. Example: If you’re forecasting the chance of a new business succeeding, start by asking, “What percentage of new businesses typically succeed within five years?” This is your baseline. Why it works: Base rates provide a reality check, showing how common an event is within a larger class of similar events.
3. The Inside View: Add Your Unique Insight
Once you have the outside view, it’s time to factor in specifics. The inside view is where your hunches, assumptions, and unique context come into play. How to balance it: Look at the unique circumstances: What makes this situation different? Adjust your prediction based on these nuances. Avoid letting your adjustments deviate too far from the base rate. Example: If the new business is led by an experienced entrepreneur, you might increase its odds of success slightly above the baseline.
4. Thesis, Antithesis, and Synthesis: Thinking in Layers
This technique comes from philosophy but applies perfectly to forecasting. It’s about considering multiple perspectives and combining them into a well-rounded conclusion. Steps: Thesis: State your initial prediction. Antithesis: Challenge it by asking, “What if I’m wrong?” Synthesis: Reconcile the two perspectives to create a more robust prediction. Why it’s useful: It forces you to confront your biases and refine your forecast.
5. The Crowd Within: Two Heads Are Better Than One (Even if Both Are Yours)
This technique involves generating two independent estimates and then merging them for a more reliable prediction. Steps: Make your first estimate based on your initial thoughts. Assume this is wrong, and challenge yourself to create a second estimate from a different angle. Combine the two: Average them or take a weighted approach. Example: Predicting election results? Your first estimate might use polling data, while the second considers economic trends. Merge them for balance.
6. Dialectic Discussions: Feedback Is Key
Engage in thoughtful discussions with others to refine your forecasts. This helps you avoid groupthink and leverage diverse perspectives. How to do it: Share your initial prediction with someone else. Ask for their reasoning and feedback. Compare your thoughts and adjust your forecast accordingly. Tip: Seek out people who challenge your views rather than just agreeing with you.
7. Pre-Mortem Analysis: Spot Problems Before They Happen
A pre-mortem involves imagining that your prediction has already failed and then brainstorming reasons why. Steps: Assume your forecast turned out wrong. Ask yourself, “What went wrong?” List potential pitfalls and adjust your prediction to account for them. Why it’s useful: It helps you anticipate and mitigate risks before committing to a forecast.
8. Probabilistic Thinking: Quantify Your Confidence
Forecasting isn’t just about making predictions—it’s about assigning probabilities to outcomes. How to do it: Use percentages to express your confidence (e.g., “I’m 70% sure this will happen”). Regularly update your probabilities as new information becomes available. Avoid being overconfident by considering a range of outcomes. Example: Instead of saying, “The stock market will rise,” say, “There’s a 60% chance the market will rise in the next quarter.”
Final Thoughts: Practice Makes Perfect
Becoming a great forecaster takes time, but these techniques will set you on the right path. Start with small questions, practice regularly, and always be open to learning and improving. Happy forecasting!