Tetlock and Dan Gardner, his co-author on Superforecasting:
The Art and Science of Prediction, set out to narrow down the ten main ideas that have been “experimentally demonstrated to boost accuracy” in the real world.
Pay attention to the questions where your effort will likely be rewarded. Avoid wasting time on questions that are simple and “clocklike” (where common sense can help you get close to the correct answer) or difficult and “cloudlike” (where even sophisticated statistical models fall short against the dart-throwing chimp). Focus on questions that fall within the Goldilocks range of difficulty, where effort is most beneficial.
- Son is Not Alone. Millions Young
- Stanley Cup Finals Canadiens Still Out to Prove They Belong
- Stop. Breathe. We Keep Like This.
Don’t inquire, “Who will win the World Series in 2050,” for instance. That is unpredictable and unknowable. As we get closer to home, the question gets more intriguing. It is a different question to ask in April who will win the World Series for the upcoming season and how much justifiable confidence we can place in that prediction. Even though we don’t have much faith in the outcome, we can do much more than try to guess who will win in 2050. At worst, we can reduce the possibilities. This enables us to return to the risk-uncertainty continuum.
A well-deserved reputation for being radically unpredictable exists for some classes of outcomes (e.g., oil prices, currency markets). However, we frequently don’t realise how unpredictable the results are until we’ve tried in vain to advance our analytical understanding. Remember the two most common mistakes you could make in this situation. Either we could waste time trying to predict the unpredictable or we could fail to try to predict the potentially predictable. Which mistake would be more detrimental given the circumstances?
2. Disassemble Issues
This is thinking in the Fermi sense. The initial atomic reactor was created by Enrico Fermi. When he wasn’t doing that, he enjoyed answering difficult inquiries like, “How many piano tuners are there in Chicago?”
This initially appears to be very challenging. Fermi began by breaking the issue down into smaller components and categorising them as either knowable or unknowable. By approaching a problem in this way, you can reveal your ignorance or, in the words of Tetlock and Gardner, “flush ignorance into the open.”
It is preferable to expose your presumptions and identify your mistakes quickly than to conceal them behind jargon and obscurity. Fermi-izing is a skill that superforecasters excel at, even when it comes to seemingly irrational things like love.
The surprise is how frequently remarkably accurate probability estimates come from an incredibly rudimentary set of presumptions and educated guesses.
3. Equalize Views From Inside and Outside
Tetlock and Gardner add a historical perspective, echoing Michael Mauboussin who cautioned that we should pay attention to what is the same:
Superpredictors are aware that nothing new can be discovered. There is nothing truly “unique.” Be damned, linguistic purists: uniqueness is a matter of degree. So even for events that seem to be out of the ordinary, such as the outcome of the manhunt for high-profile terrorist Joseph Kony or the standoff between a new socialist government in Athens and Greece’s creditors, superforecasters conduct imaginative searches for comparison classes.
How frequently do events of this nature occur in circumstances of this nature? is a question that superforecasters frequently ask.
- Healing is TikToks Latest Trend. Does
- Olympic Skateboarding Horigome Soars and Huston Stumbles
- Fistful of Vengeance Review Pursuing the Queen of the Underworld
4. Refresh Your Convictions
As brushing and flossing are to good dental hygiene, belief updating is to good forecasting. While occasionally tedious and uncomfortable, it eventually pays off. Despite this, don’t assume that updating beliefs is always simple just because it can be. Teasing out nuanced signals from chaotic news flows while avoiding the allure of wishful thinking is necessary for skillful updating.
Smart forecasters develop the ability to spot warning signs before the rest of us do. They look for non-obvious lead indicators, such as what would need to occur before X, which could be anything from an increase in Arctic sea ice to a nuclear conflict on the Korean peninsula. Here, it’s important to note the fine line between being the first to spot subtle clues and being duped by deceptive clues.