Updating your beliefs can be boring or uncomfortable but it very beneficial in the long run. Look for subtle signals in noise and avoid wishful thinking. The world doesn't work the way we want it so ideally we would behave like rational Bayesian updaters.

University of Michigan - Ann Arbor

Philip Tetlock and Dan Gardner promote 10 key themes that have been experimentally demonstrated to improve prediction accuracy. These themes are:
- 1) Triage
- 2) Breaking problems down
- 3) Balance Inside and Outside Views
- 4) Update your Beliefs
- 5) Everything is Connected
- 6) Remove Uncertainty
- 7) Balance Prudence and Decisiveness
- 8) Learn from Failure and Success
- 9) Manage the team
- 10) Master the error-balancing bicycle



10 themes for Superforecasting

Created by Thomas Bayes, who posited that our estimate of the probability of an outcome was determined by two things:
- 1) prior probability of an outcome 
- 2) likelihood of an outcome

Bayesian Inference

https://fs.blog/2015/12/ten-commandments-for-superforecasters/

Ten Commandments for Aspiring Superforecasters

It is better to invest time in answering questions where your effort is likely to pay off. This means finding a balance between easy-to-answer and impossible questions. 

Triage 

Nothing is completely unique, there are only degrees of originality. Even in seemingly unique events, it is worth conducting creative comparative searches. It can be useful to pose out-side view questions such as "How often do things of this sort happen in these situations?"

Balance Inside and Outside Views

Update your Beliefs

For every good argument there is an extremely likely counterargument that is at the least worth acknowledging. You should be willing to be nudged by the other side just as you would nudge them. Synthesis reconciles irreducible, subjective judgements. It is usually better to have a nuanced perspective.

Everything is Connected

Nuance matters, you should distinguish as many degrees as possible. For example, you will have an advantage in poker if you know what your odds are for different hands.

Remove Uncertainty ( in prediction )

You do not want to rush to judgement or take too long. You should think about the trade-off between taking decisive stands and validating that stand. Improving performance takes good scores on calibration and resolution.

Balance Prudence and Decisiveness

Never justify / rationalize your failures. Analyze where you went wrong to determine how to get better (also recognize when you were just lucky).

Learn from Failure and Sucess

Team management is important, especially leveraging perspectives: 
- People should be able to reproduce others argument
- Help others clarify their arguments through questions
- Be constructive with confrontation, disagree without being disagreeable

Manage the Team

To become a better forecaster, you must balance opposing errors to strike an optimal strategy. Learning requires doing (with feedback), not just reading. 

Master the error-balancing bicycle

The task decomposition heuristic is a problem-solving strategy that involves breaking down a large or complex goal into a series of smaller, more manageable steps. This approach makes overwhelming tasks feel more achievable. It also helps in identifying errors and separating known elements from unknown ones, a technique famously used by physicist Enrico Fermi.

Task Decomposition Heuristic

The likelihood of an outcome represented by the extent to which available evidence points to the outcome.

Likelihood (Bayesian Inference)

Commonly referred to as the prior, it is the prior probability of an outcome occurring, represented by our initial belief about the probability of an outcome.

Prior Probability (Bayesian Inference)

Bayesian networks were developed to understand reasoning under conditions of uncertainty. It is based on Bayes Theorem and is a probabilistic model that represents a set of variables and their conditional dependencies by being able to find the probability of facts being true/false given already defined facts. A Bayesian network is represented by a directed acyclic graph which helps to visualize the probabilistic model for reviewing relationships between random variables and reasonings for causal probabilities given certain facts



Learn Before

Related