Predictions and forecasts are made about the weather, the markets, the future job and political outlook. They have entered into most areas of our lives, whether we’re mapping out our next career move or choosing where to invest, leading to our fixation on them and our irritation when things don’t go as predicted.
In their book ‘Superforecasting: The Art and Science of Prediction’, Tetlock and his colleague Dan Gardner found that forecasters could improve their accuracy by studying the data and a few other best practices. They found that it’s possible for forecasts to upgrade their skills, realign with each new piece of information and then analyzed and improved upon. Superforecasting is a trainable skill that everyone can do with some evidence gathering, score keeping, and patience.
Key Points From Superforecasting
- Prediction is a talent and a skill that can be developed. Anyone with enough dedication, interest and domain expertise can improve their skill and accuracy
- Our complex world means that small events can lead to large unforeseen consequences making regular forecasting rather limited. This, however, does not mean that forecasting should be scrapped. Just because not everything is predictable, doesn’t mean everything is unpredictable.
- Superfocasting is derived from curiosity, the desire to learn, the ability to gather information, and the willingness to change and update our beliefs.
- Superforecasters are less ideologically (or professionally) biased. They seek data from a wide range of sources to examine future trends. They are open-minded and have less fear of making mistakes.
- Superforecasters embrace probabilistic thinking and understand how statistical probability works.
- As forecasters, we want to measure the accuracy of our forecasts in order to improve our forecasting skills. We also want to adjust your forecast as new information comes to light.
- Forecasters should avoid using vague language like “might”, “could” and “likely” because different people attach different meanings to these words, it’s far better to use numbers, particularly percentages.
- Seemingly impossible forecast problems can be tackled by breaking them down into smaller bite-size units to analyze
- Every situation is unique so don’t judge a case too quickly. Approach it from the outside by finding the base rate first.
- It’s always wise to plan for adaptability and resilience.
- Many forecasters tend to give advice that is too certain due to the fact that consumers have an inherent distaste for uncertainty
- Declarations of high confidence mainly tell you that an individual has constructed a coherent story in his mind, not necessarily that the story is true.
- In forecasting, often, a small edge can make a big difference.
- The more that is unknown, the greater the opportunity.
- Consensus is not always good; Disagreement not always bad. If you do happen to agree, don’t take that agreement—in itself—as proof that you are right. Never stop doubting.
- Beliefs are hypotheses to be tested, not treasures to be guarded.
- We as humans have the ability to take a series of past events and turn them into linear narrative that makes the outcome seem all but inevitable.
- Using multiple lenses to view a subject can provide a more detailed picture and yield greater understanding and better forecasting.
- One of the keys for better forecasting is to embrace diverse thoughts, expose assumptions, catch mistakes, and correct biases.
- People who work in teams had forecasts that were more accurate than forecasts made by individuals. However, it's key to include critical discussions and present an alternative view.
- The goal of forecasting is not to see what’s coming. It’s to advance the interests of the forecaster and the forecaster’s community.
- Anchoring bias: “When we make estimates, we tend to start with some number and adjust. The number we start with is called the anchor. It is important because we typically underadjust, which means a bad anchor can easily produce a bad estimate.”
Measure and Update Your Forecasts
Weather forecasts are quite reliable when made a few days in advance because forecasters analyze the accuracy of their forecasts with the actual weather that has occurred, this allows them to improve their forecasts alongside their understanding of how the weather works. Unfortunately, forecasters in other fields do not usually measure the accuracy of their forecasts so in order to improve forecasting we have to get serious not only about comparing but about measuring and updating our findings.
You can also use the Brier score system to see how accurate past forecasts were. The Brier score system states that the lower the score, the more accurate the forecast so if the forecast was perfect, it gets 0. Forecasts which are randomly guessed are given a score of 0.5 whilst completely wrong forecasts are given a maximum score of 2.0. Interpreting the score depends on the question being asked as you might have a Brier score of 0.2 which seems great yet the forecast could be terrible.
Favorite Quotes from Superforecasting
"The more that is unknown, the greater the opportunity."
“For superforecasters, beliefs are hypotheses to be tested, not treasures to be guarded.”
Get actionable takeaways and self-exploration exercises of one of my top recommended books of all times. Just enter your email below and it will be sent to you instantly for free.
* No fluff, No spam, You can always opt out
* No fluff, No spam, You can always opt out
Editor and Founder
Tal Gur is a location independent entrepreneur, author, and impact investor. After trading his daily grind for a life of his own daring design, he spent a decade pursuing 100 major life goals around the globe. His most recent book and bestseller, The Art of Fully Living - 1 Man, 10 Years, 100 Life Goals Around the World, has set the stage for his new mission: elevating the next generation of leaders to their true potential.