Suggest edit — Principles of Decision Making

Title

Name

Note

---
title: "Principles of Decision Making"
visibility: public
---

# Principles of Decision Making

Category: [[decision-making|Decision Making]]

[Read the original document](https://docs.google.com/document/d/1wW81B7g-bKd0km9ey6d-Ynpz3Rzx9Ie5sftpbDX-SPw/edit?usp=sharing&sa=D&ust=1596495076357000&usg=AOvVaw3WxdFcnLgAb94M_SgSAuQU)

---

A Synthesis of the Decision Making Literature.

Motivation
For summaries of the frameworks presented, see the linked documents below for details on Principles and Decisive.

Both the Sequences and Thinking Fast and Slow provide an overview of how thinking, beliefs, and intuitions can be systematically wrong and describe some ways in which those incorrect beliefs lead to poor decisions. The downside of both is their actionability. Thinking fast and slow is oppressively abstract in places. The Sequences is relatively disorganized, and while it discusses challenging irrational thinking it is certainly light on explicit advice for deciding.

Baron’s Thinking and Deciding is also relatively heavy on the thinking and light on the deciding. It is a great introduction to expected utility theory, theory on rational belief formation, major concepts like opportunity costs, values, and retrospective decision analysis. It is rather technical and so deals well with decisions that can be formally specified. Some of the examples are explicated in detail. 
A slightly less technical, substantially shorter and more practical version of this book would be my decision making bible. I do not feel comfortable recommending baron as all but the most serious and intense readers would be frustrated by the investment it requires.

Section 5 of life principles (Learn how to make decisions effectively) in Dalio’s Principles is the closest to what I’m looking for that I’ve found to date. It’s a 30 page tour of emotional management, the tradeoff between information acquisition (learning) and decisiveness, decision making at the right level of analysis (ex., big picture vs. operational decisions) and a few other concepts. It misses many important concepts (ex. goal factoring or its equivalent) but if I was looking for the read that gave me 80% of the content in 20% of the time, this would be it. I’m looking for ~99% of the content with 100% of the time.

Decisive generated a lot of useful leading questions to inform each decision.

My major frustration is that none of these seem to give solid answers to triage - how to cut down a large space of important decisions, all of which are incredibly complex, to a very small set of decisions that accord with your resources.

My research decisions, writing decisions, project choices, mentor and mentorship choices, and social decisions have all become much more complicated in the past 2 years. It’s very hard for me to tell you with any confidence whether many of my decisions have been correct. I feel like I’ve hit what got you here won’t get you there, as old decision processes clearly bore enormous fruit while the same processes struggle today.

I sense that the complexity of my decision making has increased dramatically, because decisions that used to have competing options with clear winners now have more competing options which are all compelling in their own way and seem to realize a world aligned with my values through very different paths.

It’s much harder to correctly predict what decisions will succeed, harder to evaluate the counterfactual case and so know whether past decisions were good or bad, and in general information that is relevant to my decisions has become much more niche and context dependent and so has been much harder to come by.

Decision Making Processes:
https://docs.google.com/document/d/1f0DiLJkjE8UdjUxQRl3sjUfTZGyHg6eOXT7-m_vdHxE/edit?usp=sharing

Compressed Concepts in Decision Making:
https://docs.google.com/document/d/1bK_hikOIsWwKutx6D3psO2pvtP-O3r5GDjcwDPCSq8c/edit?usp=sharing

Implementing Principles:
https://docs.google.com/document/d/1OKqQKPhnAQFI0-CQ32uqoCp1DilyDbQDusTqLHJeXO0/edit?usp=sharing

Notes Principles:
https://docs.google.com/document/d/1LeCX4z2RX2lTwct1Ew93mgJZhHQHZ07MxD60y0EFMu4/edit?usp=sharing

Decisive:
https://docs.google.com/document/d/1KI6QvbUDGgcCA9DiWsGIzci4ya5ngGWasy2f7Jh5l2Q/edit?usp=sharing

Decisive Notes:
https://docs.google.com/document/d/1yWqgE2HyZ4dplhenF9mSZHkG3nyeqD44Cik8_XUw_vw/edit?usp=sharing

Decision Making Techniques and Models
Major Techniques
1. Go to underlying reasons for the decision, and reason up. Goal Factor.
2. When possible, run a trial of important decisions before committing to them.
3. Define and enshrine core priorities & values.
4. Premortem
5. Find objective Information. 
6. Zero Based Thinking
   1. If a new person walked into your life, what would they do? Escape your mental frame.
   2. If I was not already in this job / relationship / situation, would I enter it again?
   3. If any variant of no, ask: How do I get out, and how quickly can I do it?
7. Distance yourself emotionally from the decision
   1. Worst case scenario
8. Minimize the downside of your decision
9. Set tripwires to check whether a decision needs to be made / changed
10. Assume you can’t choose any of the existing options. What would you do?
11. What would the most determined, courageous version of me do?
12. Assume you’re forced into one decision. What would you do?
13. Have multiple good options in front of you simultaneously (Protects from premature commitment)
14. Seriously consider the opposite of important decisions. Find a belief driving your behavior and argue with all energy against it.
15. Look at your decision from the perspective of several time frames. 10 days / 10 months / 10 years, for example.
16. When you have a difficult decision to make, flesh it out as the values conflict it inevitably is.
17. Honestly ask yourself what you want and what should be done about it.
18. What am I likely to lie to myself about in this space?

Models
1. Learning from Failure
2. Facing Reality vs. Pretending / Ignoring Reality
3. Determination / Courage
4. Social Norms
5. Opportunity Cost
6. Triage / Prioritization (80/20 Pareto Principle)
7. Confirmation Bias
8. Short term emotions
9. Prevention Focus vs. Promotion Focus
10. Identify and don’t tolerate the problems that stand in the way of your goals
11. Design plans that explicitly lay out the tasks that will get you over your problems and onto your goals

Literature Review
In each case, adding the place that the best resources cover the concept. I may distinguish between useful techniques for decision making and useful ways of thinking about decision making, but ideally I’d have a tagging system that let me tag these.
Decisive. Chip Heath & Dan Heath.

* [Pg. 18-23] Four Villains of Decision Making - There are four major steps in a decision making process. Each step has a ‘villian’.
   * Stage One: You encounter a choice.
      * Villain: Narrow framing makes you miss options.
      * Solution: Widen your options. Expand your set of choices. [Systematizing Creativity is entirely about this kind of solution]
   * Stage Two: You analyze your options.
      * Villain: Confirmation bias leads you to gather self-serving information.
      * Solution: Reality-test your assumptions. Get outside your head and collect information that you can trust.
   * Stage Three: You make a choice.
      * Villain: Short term emotion tempts you to make the wrong choice.
      * Solution: Attain distance before deciding. Asking a long-term question like ‘what would our successors have us do?’ can help dramatically.
   * Stage Four: You live with your decision.
      * Villain: You’ll be overconfident about how the future will unfold.
      * Solution: Prepare to be wrong. Plan for an uncertain future. [The Strategy Paradox is entirely about this kind of solution]
* [Pg. 29] You, by default, form an opinion on all things. Danny Kahneman: “A remarkable aspect of your mental life is that you are rarely stumped.”
* [Pg. 29] Process matters much more than analysis. [I think this is an incredibly underserved point if true. Decision processes are incredibly rare outside of some corners of business.] Source: Study by Lovallo and Sibony called ‘The Case for Behavioral Strategy’. They study 1048 real world decisions about products, mergers and acquisitions, and large capital expenses. They ask about 17 practices applied in making the decision. Eight of them are about the quality of the information gathering & analysis. The other nine are about the process used to make the decision. [McKinsey Summary]
   * Ex., ‘Did you explicitly explore and discuss major uncertainties or discuss viewpoints that contradicted the senior leader’s?’
* [Pg. 29] Against pros-and-cons processes.
* [Pg. x] Most of the decision making literature usus a glorified spreadsheet where you identify your options, choose a few key axes, weights for those axes, and put measures to each possible decision on each axis and trust the math.
* [Pg. 49] Spotlight effects get us stuck in a narrow frame, focusing on our current options.
* [Pg. 49] Think about opportunity cost to avoid being trapped in a narrow frame.
* [Pg. 49] Ask, ‘What if your current options disappeared?’ to move the spotlight to a new possibility.
* [Pg. 67] Multitrack - consider more than one option simultaneously. This lets you learn the shape of the problem. ‘This and that’ rather than ‘this or that’.
* [Pg. 67] Decision paralysis becomes a concern for people who consider many options, and so the authors push for only one or two extra. The payoff can be huge for just the first or second additional option. [Note - rather than providing a solution to decision paralysis in the presence of too many options, they focus on providing a very small number of additional options. This may indicate that they believe that decision paralysis in the face of many options is unsolved.]
* [Pg. 67] Move between prevention and promotion mindsets. Prevention focus is about avoiding negative outcomes. Promotion focus is about pursuing positive outcomes. Using both mindsets is far superior than just one or the other.
* [Pg. 68] Find someone who’s solved your problem.
* Leading Questions:
   * Testing is much better than prediction. Can I do an experiment?
   * What are the options that we’re not considering?
   * What objective information should inform our decision?
   * What beliefs do I have that I should question?
   * What is the evidence that contradicts my beliefs?
   * What are the underlying desires guiding my decision?
   * Assume that some or all options are eliminated or required. What do we do?
   * What is the opportunity cost of this decision?
   * Can we reframe between prevention focus and promotion focus?
   * Who else had this problem, and how can we learn from them?
   * What evidence, if witnessed, would change your mind?
   * When this whet well, what was happening?
   * What was the close-up experience of people who made my decision in the past?
   * How will I feel about this decision in 5 minutes, in 4 months, in 3 years?
   * Premortem - imagine this decision going horribly wrong and going extremely well. What caused it?
   * What is the courageous action? The imaginative action? 
Principles. Ray Dalio.

*  [pg. 236] Decision-making happens in two ordered steps: learning, then deciding.
   * Learning needs hyperrealism and radical open-mindedness.
      * Learning may require solving some emotional blocks.
      * Quality of learning is based on the ability to synthesize accurately and knowing how to navigate levels.
   * Once you’ve learned enough, you can decide. Deciding involves examining first, second, and third-order consequences of your decision.
   * Avoid the reverse: deciding, then learning. Avoid subconsciously picking your decision first, then cherry-picking data to support it.
* [pg. 237] Synthesize the situation 
   * Filter details through a high-level perspective.
   * Carefully examine the people you ask questions of. Having no answers is better than listening to the answers of uninformed people.
   * Screen all inputs for believability.
   * Keep events in a long time-scale perspective.
   * Beware novelty bias. It is smarter to choose great over new.
   * Know when collecting more data beats analyzing existing data.
* [pg. 239] Synthesize the situation over a long timescale 
   * Decompose what improvement means. Hold in mind rates of change, levels, and interrelationships for the metrics you care about. E.g. instead of “it’s getting better,” what is the rate of change, and will that rate of change get you above the bar in an acceptable amount of time? 
   * Making good approximations is very undervalued in a culture obsessed with precision. Beware of lengthy discussions about the exceptions, forgetting about the rule.
   * Know the key 20% in the 80/20 principle. Beware marginal returns.
   * Be an imperfectionist. Isolate the 5-10 most important factors instead of focusing on every detail.
*  [pg 247] Effectively navigate levels
   * Create an outline mapping your high-level values to your everyday behaviors.
   * Pay attention to your conversations, and how you move between levels as you talk.
   * Label conversations as “above-the-line” and “below-the-line.” Above-the-line conversations focus on high-level details and only go below-the-line to illustrate a main point.
   * Decisions should be made at one level, but consistent across all levels.
* [pg 251] Two levels of decision making: evolution/logic-based and subconscious/emotion-based. Beware the latter.
   * Quote from Carl Jung: “Until you make the unconscious conscious, it will direct your life and you will call it fate.” 
   * Use logic, reason, commonsense, evidence-based decision-making:
      * Use expected value calculations: the reward times the probability of occurring is greater than the cost times the probability of occuring. [Note: what about evolutionary biases that make us fear loss more than want gain? How do we reconcile our biases with being rational actors? And may these evolutionary biases be more optimal in important contexts?]
      * Always try to raise the probability that you are right in your expected-value calculation. 
      * Know when not to bet.
      * The best choices may have some cons, versus no cons at all. Be careful of people who argue against something if they can find any negative feature.
* [pg 254] Prioritize by weighing the value of additional information against the cost of not deciding.
   * Constantly evaluate the marginal benefit of gathering more information against the marginal cost of waiting to decide.
   * Separate “must-dos” and “like-to-dos” and do the former first. Be careful not to slip the latter into the former.
   * Probabilities matter more than possibilities. Don’t be a “philosopher type” who confuses the two LOL.
* [pg 255] Simplifying is harder than complicating. Practice simplifying effectively.
* [pg 255] Use principles / abstractions. [Note: Oh great! he has a process of abstracting from your thoughtstream:]
   * Slow down your thinking so you can note the criteria you are using to make the decision.
   * Write the criteria down as a principle
   * Think about the criteria when you have an outcome to access, and refine them before the next “one of those” comes along
* [pg 256] Triangulate your decision-making with thoughtful people who aren’t afraid to disagree with you.
* [pg 257] Make decisions in partnership with a computer  [Note: This is the vaguest section so far. Does he mean literal algorithms, spreadsheets, etc? Could he provide examples?]
   * Note: he provides a brief section on computerized decision making (and his app is probs a follow up on this). But this section is actually quite important and could be populated with examples.
* [pg 257] The rest of this is describing his position on man-machine merge as a good thing. He goes into some nuances of the strengths and weaknesses of each system, but I’d rather hear more about this topic from an expert on AI interpretability. 
   * Computers are beneficial in their persistence, impartiality, and ability to process large amounts of data.
   * Humans should always understand cause and effect relationships and theories too deep to currently implement in computers.

Notes/Questions:
* Dalio’s principles is clear and comprehensive. What is missing? I feel slight frustration that he’s not saying anything especially weird, counterintuitive, or contrarian (the Sequences does this well). His book is well-distilled “commonsense.” 
   * I can’t tell if this it’s the emotional valence, the content itself, both, or neither.
* I am slightly emotionally reacting against his aesthetic. I think it’s because his tone is prescriptive. I feel like a rando Bridgewater employee.  In contrast the Sequences make you feel like a genius insider.
* Dalio’s also sensitive to the personal qualities of bad decision-makers. I appreciate this. He includes phrases that are warning flags for bad-decision making (e.g. “Wouldn’t it be good to do… X?” said by people who care about unimportant details)
* There’s a Principles app! It has a library of common situations.
* I question Dalio’s “radical honesty.” Radical honesty seems to work only when you’re already in a position of safety. But I haven’t read the radical honesty chapter thoroughly, so perhaps he addresses this.

Thinking and Deciding. Jonathan Baron.
Focus will be on
Ch. 11 (Descriptive theory of choice under uncertainty),
Ch. 14 (Decision analysis and values) and 
Ch. 19 (Decisions about the future).

* [Pg. 341] Separate utility into multiple attributes. Each attribute corresponds to a goal or value that is separate from the other attributes. Allows us to consider all relevant goals, simultaneously. Multi-attribute utility theory. 
* [Pg. 343] Discovering Values. Often, we need to discover the values worth considering. There is a book on this - Values Focused Thinking by Ralph Kenney. This tep often leads to solutions to practical problems.
   * Make a which list - If you have no limitations at all, what would your objectives be.
   * Consider the advantages and disadvantages of alternative options. These usually correspond to objectives.
   * Look for problems and shortcomings of current alternatives. This also discovers values.
   * Think of consequences. Consider why they might be good, bad, acceptable or unacceptable.
   * Look for goals (achieved or not achieved), constraints, and guidelines. These are all pointers to values. 
   * Consider different perspectives on the problem. What would someone else think?
   * Think about your own strategic objectives, the long-run fundamental values for the type of decision at issue.
   * Think of generic objectives. Organize them into means-ends relationships and into a fundamental objectives hierarchy.
   * Quantify objectives. Often this will lead to the separation of one objective into tow. Ex. both optimizing for speed and safety.
* [Pg. 360] The framework taught here, multiple attribute utility theory, is seldom used by the medical and business school students and military officers to whom it is taught. 
* [Pg. 361] Teaching is easier when the focus is on types of errors like single-mindedness (neglecting relevant goals), impulsiveness (failure to consider alternatives and evidence), and neglect of probability.
* [Pg. 475] Good reasons to stick to plans:
   * Working with other people consistently leads to a dependable reputation.
   * Plans build up their own momentum
   * If you fail, you lose faith in your ability to make plans and keep them.
* [Pg. 476] Bad reasons to stick to plans:
   * Status quo bias.
   * Omission bias (being concerned about possible harms from action than from inaction)
   * Overweighting downside risk, as in prospect theory.
   * Sunk-cost effects.
   * [Pg. 477] Commitment effects - proving yourself right becomes more important that making the right decision, and so you double down on what you’ve selected in the past.
* [Pg. 478] Discounting, and hyperbolic discounting where you discount the future much more than the short term. You are ner sighted, temporally myopic. 
* [Pg. 488] Self control as a way to appropriately value the future. Binding yourself to a course of behavior.
   * Control of emotion
   * Control of attention
   * Personal rules
   * Extrapsychic devices (removing choices from your environment)
* [Pg. 493] Experience establishes a reference point, against which future experiences are compared.

Thinking, Fast and Slow. Daniel Kahneman.
Rationality, from AI to Zombies. Eliezer Yudkowski.
Sections:
Seeing with Fresh Eyes sequence, Pg. 365
Challenging the Difficult, Pg. 1605

Poor Charlie’s Almanak. Charlie Munger.
CFAR Workbook. CFAR.
Judgement in Managerial Decision Making

* [Pg. 2] Decision Anatomy:
   * Define the problem.
   * Identify the criteria (most decisions require you accomplish more than one objective)
   * Weight the criteria by their relative importance.
   * Generate alternative courses of action.
   * Rate each alternative on each criterion.
   * Compute the optimal decision. Multiply each criterion by its weight and sum the ratings for each alternative.
* [Pg. 7-10] Biases that stand in the way of rational decision making:
   * The Availability Heuristic. Whether a cause of an event is available in memory determines our assessment of its frequency and probability. Events that evoke emotions and are vivid, easily imagined, and specific are more available than bland, vague or difficult to imagine unemotional events.
   * The Representativeness Heuristic.
   * Positive Hypothesis testing.
   * The Affect Heuristic.
* [Pg. 11] Major Principles
   * Common Biases.
   * Bounded awareness.
   * Framing, perceptions of change, and reversals of preference.
   * Motivation and emotion.
   * Escalation of commitment.
   * Fairness and ethics in decision making.
   * Contexts:
      * Common investment mistakes.
      * Making rational decisions in negotiation.
   * Six strategies for improved cognition:
      * Use prescriptive decision making procedures.
      * Acquire expertise.
      * Debias your judgement.
      * Reason analogically. [Seriously? Or is this reference class forecasting?]
      * Take an outsider’s view.
      * Understand biases in others.

Notes:
This book is inspiring in its simplicity.
I’d like to point their decision anatomy at my biggest high level problems and goals.

The Effective Executive.

Ch. 6 (The Elements of Decision Making) and 7 (The Effective Decision).

* [Pg. 113] Executives do not make a great many decisions. They concentrate on the important ones.They try to think through what is strategic and general, rather than “solve problems”. They try to make the few important decisions on the highest level of conceptual understanding. They want impact rather than technique, they want to be sound rather than clever.
* [Pg. 114] Effective executives know when a decions has to be based on principle and when it should be made on the merits of the case and pragmatically. They know that the trickiest decision is that between the right and the wrong compromise and have learned to tell one from the other.
* [Pg. 114] The most time-consuming step in the process is not making the decisio nbut putting it into effect.
* [Pg. 122] The Elements of the Decision Making Process:
   * The realization of whether the problem is generic can be solved through a decision that establishes a general rule or principle
   * The development of the specifications which the answer to the problem has to specify.
   * Thinking through the solution that will absolutely satisfy the specifications, before giving attention to the compromises, adaptations and concessions needed to make the decision acceptable.
   * The building into the decision of the actions to carry it out.
   * The feedback which tests the validity and effectiveness of the decision against the actual course of events. 
* [Pg. 144] A decision is a choice between alternatives. A judgement. It is often a choice between ‘almost right’ and ‘probably wrong’ - a choice between two courses of action which are not provably better than one another.
* [Pg. 145] Emphasis on change of measurement in decision making.

Decision Theory: Principles and Approaches. Giovanni Parmigiani. Lurdes Inoue.
The Model Thinker. Scott Page.
Super Thinking. Gabriel Weinberg. Lauren McCann.
Tempo. Venkatesh Rao.
Systematizing Creativity (Plan Generation)

Generating novel plans is a critical prerequisite to making dramatically better decisions.

Methods

1. Intentionally enter diffuse mode over ideas
2. Abstract and Generalize / Transfer over similar problems & solutions
3. Composition / Recombination
4. Idea Lists
5. Decomposition
6. Randomness
7. Idea Mapping, Graphs of Relationships between Ideas
8. Leading questions
9. Reframe / Question Assumptions
10. Multiple levels of analysis
11. Think ground up, from first principles
12. Automation
13. Thought Habits / Mental.
14. Invert
15. Activities 
16. Social Solutions
17. Other

Implementing Methods

1. Intentionally enter default mode over ideas
   1. Load up an idea / problem / question and:
      1. Go for a walk
      2. Take a shower
      3. Sleep
         1. Start to fall asleep, wake up as you do (use alarm or rock in hand)
         2. Hypnagogia
      4. Meditation
         1. Sit in silence with the target as an object
2. Abstract and Generalize / Transfer over similar problems & solutions
   1. Model vs. Technique - see what works in the space, ask why to get a model. Generalize from the model to generate more techniques.
   2. Metaphor Generation / Analogizing
      1. Idea list over metaphors for a given problem / solution / object
      2. Transfer solutions and insights from the related domains
   3. Find a source idea, categorize it, generalize to finding more instances of that category.
      1. Ex. Properties of Representation, Systematizing Creativity
   4. List solutions to a problem and generalize
   5. List related problems and generalize from their solutions4400666544191253
3. Composition / Recombination
   1. Idea List for concept set
   2. Run recombination over generated concepts
      1. Hold the concepts in mind, asking how they relate to one another.
4. Idea Lists
   1. List Creation + Time Pressure
      1. Choose a topic / prompt / question.
      2. 10m Time Constraint. 
      3. Fill list to 10 ideas. 
      4. If time becomes a limiting factor, let novelty / quality fall.
   2. Alternative versions:
      1. Go into diffuse mode over an idea list, with no time limit
      2. Create a huge list (with a low barrier to idea entry) and prune it
         1. This resolves the psychological conflict between creative ideation and rigor / quality
5. Decomposition (Mapping out the space)
   1. Break into component pieces, in multiple directions
      1. Ex. Machine learning becomes Linear Algebra + Calculus + Probability Theory + Computer Science, which break into their own subregions
      2. Ex. Scientific Field becomes Major Papers + Categories of the topic + Conferences + Major Researchers + Quality Sites
   2. Mutually Exclusive, Collectively Exhaustive
   3. Deconstruction + Optimization
   4. Actually do science ‘to split’
6. Randomness
   1. Randomize. Generate random ideas by specifying some parameters, and make them work / use them as prompts.
   2. Randomly show words that serve as prompts
   3. Stream of Consciousness
7. Idea Mapping, Graphs of Relationships between Ideas
   1. Create a graph of the relationships between critical ideas in a space
8. Leading questions
   1. Recursive ‘Why?’
   2. Questions over answers
   3. Imagine the future (problem is solved, for ex.). What happened? Work backwards.
   4. What are the sacred beliefs? What can’t be thought?
   5. "what if" questions
   6. "how might we" questions
   7. Invert - "what if the opposite is true"
   8. Eliminate - “does it even matter”
   9. “What if I need to solve it once and for all”
   10. Scalability - “What if I need to solve it for everyone”
   11. What is the meta level idea?
   12. What questions do I have about this?
   13. What would other people think of?
9. Reframe / Question Assumptions
   1. Constraints
      1. Create resource constraints (time, attention, money, assumptions, etc.)
         1. How can I slightly modify the goal to accomplish it under insane constraint?
      2. Create resource excess (time, attention, money, etc.)
      3. Eliminating options
      4. What are the upstream constraints in the system?
      5. Define boundaries of solution spaces better
         1. Find upstream constraints
   2. What are the basic principles of x?
   3. Apply different modes of processing
      1. What would a supervillain do? (Prompt framing) / Supervillain mode
      2. Emotional - Get into emotional state and generate ideas
         1. Anger
         2. Gratefulness
         3. Adoration
         4. Frustration
         5. Excitement
      3. Types of Thinker - What would I generate if I was a:
         1. Mathematician
         2. Technologist
         3. Computer Scientist
         4. Philosopher
         5. Psychologist
         6. Economist
      4. find inspiration in other areas:
         1. math
         2. mythology
         3. writings about principles
         4. Physics
         5. Etc
      5. Environmental
         1. Work in a cluttered environment
      6. Game Lenses, list of generic lenses
      7. Asking what would Hufflepuff / Gryffindor / me would do
      8. Asking what a friend would do
      9. Predicting what someone will say and 
         1. Then asking them, for interesting feedback
   4. What? / Why? / How?
10. Multiple levels of analysis
   1. Multiple levels of abstraction - ask what alternate levels of analysis exist, through decomposition and abstraction over the current level of analysis
      1. Multiple frames - think at lower and higher levels of analysis, simultaneously. Ask how they interact.
   2. Meta-object two space
      1. Simultaneously optimizing the object and the meta level
11. Think ground up, from first principles
   1. Ask what the underlying goal is for a space, and for what solutions would serve that goal. For each solution, think of the components necessary to make that solution happen.
12. Automation [See Expanded Version]
13. Thought Habits / Mental
   1. Idea List Habitually
   2. Brainstorm Habitually
   3. Create and refine a distinct open mode
   4. Create imminent desire for coming up with relevant ideas
   5. System 1 + Generalization
      1. Take an intuitive response and understand its mechanism. Turn the mechanism into a generator.
14. Invert
   1. Take any technique, and do the opposite over some parameter
   2. Imagine ways of not doing it, or preventing the goal from being reached: adversarial
15. Activities 
   1. List Creation + Time Pressure
   2. Writing
      1. Write freely over the topic / question / prompt
   3. Brainstorm [thought dumping]
   4. Defend a difficult position, adversarial conversation
   5. Drawing
   6. Giving a speech to the air
   7. Improv Games
16. Social Solutions
   1. Crowdsourcing ideas
   2. Read Books / Articles on Topic
   3. Look up what other people have been saying about it
      1. discuss things with others
      2. check social media
      3. find differing discussions online
      4. Mapping ideas generation for other people!
   4. Work with other dissimilar people
17. Other [Could get moved up]
   1. Original seeing / childlike mind [linked to first principles / questioning assumptions]
   2. Merely observing without judgement, seeing without interpreting [linked to questioning assumptions]
      1. Allowing uncertainty to exist, not filling it prematurely with an inadequate theory
         1. This avoids confirmation bias
   3. Alternating between observing reality and creating theories / hypotheses, seeing how both are necessary.
   4. Potential adds to reframing:
      1. Noting what is absent
      2. Obsessively studying details
      3. Focus on anomalies
   5. Thinking Visually (especially about engineering and physics)
      1. To da Vinci, drawing and thinking were the same
      2. Faraday taking lessons in drawing to think visually

Underserved Topics in Decision Making
1. Systematicity
2. Actionability (ex., Spreadsheet style weighted criteria is too slow for daily, even weekly decisions) - need for quality decision processes at every time scale.
3. Cutting down many options into a few good options (triage).
   1. Standard decision methods assume all options can be evaluated on the same axes.
4. Categorizing plans.
   1. Sets of related plans that share the same direction or elements.
5. The how behind effectively generating novel options and plans.
6. The how behind first principles thinking.
7. Discovering values. Values drive goal choice, but are seldom correctly identified.
8. A substantial fraction of the value of decision processes may be due to added conviction and decisiveness, which brings gains from focus and a lowered opportunity cost.
9. The use of local heuristics vs global plans
10. The how of coming up with good numbers quickly when scoring options against criteria.
11. Decision making as future simulation.
12. Literature doesn’t cover the right personal decisions and doesn’t cover the comprehensive set of decisions that are constantly being made
13. Doesn’t interface with decision making habits, though arguably this is close to heuristics

Actionability

Make the decision making literature much more actionable, ex. Implementing Principles. One major thesis here is that our decision making tools (ex., goal factoring, premortems, being conscious of the short-term bias) work and work well, but aren’t used often enough or intuitively enough.

Upside

This would likely result in a set of incredibly compressed, runnable and trainable algorithms. It would instantiate the theory in a number of incredibly specific situations.

In practice I (and all of us) are sitting on top of an incredibly valuable and complex toolset that we never use.

The real problem, rather than generating or communicating new useful tools, is to make this tools easily used in the situation in which they’re most valuable.

Omnifocus is an example of embedding execution content directly into the system you use to operate daily.

Creativity

Demonstrate the creation of dramatically different plans that achieve the same underlying goals.

First principles thinking is incredibly important. Reframing covers it. Categorizing problems so that solutions can be transferred is incredibly important and underdone.

An Informational Perspective

Intersection of machine learning and decision making. Probability theory and decision making. The implications of insights in machine learning for thought. Focusing on the bias variance tradeoff, ensemble modeling, overfitting, hypothesis creation, regularization, training vs. generalization error, variance maximization, information gain, bayes rule (with more named consequences), exploration-exploitation. [Doc]
Bibliography

* Decisive. Chip Heath & Dan Heath.
* Principles. Ray Dalio.
   * Section 5 of Life Principles.
* Thinking and Deciding. Jonathan Baron.
* Thinking, Fast and Slow. Daniel Kahneman.
* Rationality, from AI to Zombies. Eliezer Yudkowski.
* Poor Charlie’s Almanak. Charlie Munger.
* CFAR Workbook. CFAR.
* Judgement in Managerial Decision Making. Max Bazerman. Don Moore.
* The Effective Executive.
   * Pg. 113-143.
* Decision Theory: Principles and Approaches. Giovanni Parmigiani. Lurdes Inoue.
* The Model Thinker. Scott Page.
* Super Thinking. Gabriel Weinberg. Lauren McCann.
* Boyd: The Fighter Pilot Who Changed the Art of War. Robert Coram.
   * Pg. 327-345.
* Tempo. Venkatesh Rao.

Potential Additions
* Decision Making in Action: Models and Methods. Gary Klein.
* Sources of Power. Gary Klein.
* Farsighted. Steven Johnson.
* Delusions of success: How optimism undermines executive decision making. Dan Lovalllo. Daniel Kahneman. 
* Value-Focused Thinking. Ralph Kenney.
* Superforecasting - decision making is about your predicted expected value, about the accuracy of your simulation of the consequences of each decision.

---

*Source: [Original Google Doc](https://docs.google.com/document/d/1wW81B7g-bKd0km9ey6d-Ynpz3Rzx9Ie5sftpbDX-SPw/edit?usp=sharing&sa=D&ust=1596495076357000&usg=AOvVaw3WxdFcnLgAb94M_SgSAuQU)*