Tetlock & Gardner begin Chapter 7 by using the paint-by-number method as an example for how to make good forecasts. They state that this is how forecasters unpack the components to answer questions. According to Tetlock & Gardner, forecasts are made based on the information that is available to forecasters. They identify steps that are helpful in unpacking the questions they are trying to answer by breaking down the components of the question, identify what is known and what needs to be determined, understand viewpoints from other perspectives and describe what you understand clearly and concisely. To make the best possible predictions, Tetlock & Gardner state that forecasts must be updated as soon as there is new information available to reflect the newest information provided.
They then explain that even new information can make forecasters making mistakes as they use Bill Clinton as an example in his fight to take down those who threatened America during his presidency. This leads their discussion to the concepts of overreacting and underreacting to when forecasts need to be updated. Tetlock & Gardner explain that receiving new information can be tricky because overreaction can make new information more significant than it may seem whereas underreaction can potentially not give enough credit to the new information. This happens because we want are invested in what we initially believed and do not want to consider any new evidence, especially when our initial forecasts are done in front of others. New information forces us to consider changing our opinions and beliefs which can be difficult to accept. Overacting and underreacting, as explained by Tetlock & Gardner, can both diminish accuracy and destroy good forecasts (Tetlock & Gardner 2015).
In Chapter 8, Tetlock & Gardner discuss John Keynes, an economist who became great at investing in the stock market. Keynes felt that failing gave him an opportunity to learn, identify mistakes, spot new alternatives, and try again (Tetlock & Gardner, 2015). Keynes believed that failure is a fundamental part of how we learn, progress, and increase our performance in our respective areas. This is how we improve. Tetlock & Gardner explain that consistency is a key part of our improvement. In relation to forecasting, they explain this is how forecasters become superforecasters. To become great at it you must consistently work at the skill and try. Although forecasting is a transferable skill, Tetlock & Gardner explain that it is important to practice and become better at forecasting in every area. They state that not all practice allows you to improve your skill, it must be practice that is informative of your mistakes and allows you to identify what needs improvement and more practice to get better. Identifying your mistakes is apart of learning and must be accompanied by feedback (Tetlock & Gardner, 2015). Not having feedback allows people to fall into the concept of calibration as discussed earlier in the text and can cause them to become overconfident. If people do not receive feedback, they can have a flawed view on their own performance.
In addition, feedback must be done in a timely fashion. Lagging on quality feedback allows flaws from one’s memory to creep in (Tetlock & Gardner, 2015). If we lag on time, we allow bias in on our performance. Lastly, Tetlock & Gardner explore the concept of perpetual beta. Perpetual beta is explained as the degree of which one is committed to their self-improvement (Tetlock & Gardner, 2015). They believe this mindset is more powerful than intelligence and is a quality that many superforecasters have. This concept is what predicts how great superforecasters become.
References
Tetlock, P. E., & Gardner, D. (2016). Superforecasting: The art and science of prediction. Random House.
