Collect Mental Models
Collect Mental Models
During decision-making, the brain is a memory prediction machine.
A lousy way to do memory prediction is “X happened in the past, therefore X will happen in the future.” It’s too based on specific circumstances. What you want is principles. You want mental models.
The best mental models I have found came through evolution, game theory, and Charlie Munger. Charlie Munger is Warren Buffett’s partner. Very good investor. He has tons and tons of great mental models. Author and trader Nassim Taleb has great mental models. Benjamin Franklin had great mental models. I basically load my head full of mental models. [4]
I use my tweets and other people’s tweets as maxims that help compress my own learnings and recall them. The brain space is finite—you have finite neurons—so you can almost think of these as pointers, addresses, or mnemonics to help you remember deep-seated principles where you have the underlying experience to back it up.
If you don’t have the underlying experience, then it just reads like a collection of quotes. It’s cool, it’s inspirational for a moment, maybe you’ll make a nice poster out of it. But then you forget it and move on. Mental models are really just compact ways for you to recall your own knowledge. [78]
EVOLUTION
I think a lot of modern society can be explained through evolution. One theory is civilization exists to answer the question of who gets to mate. If you look around, from a purely sexual selection perspective, sperm is abundant and eggs are scarce. It’s an allocation problem.
Literally all of the works of mankind and womankind can be traced down to people trying to solve this problem.
Evolution, thermodynamics, information theory, and complexity have explanatory and predictive power in many aspects of life. [11]
INVERSION
I don’t believe I have the ability to say what is going to work. Rather, I try to eliminate what’s not going to work. I think being successful is just about not making mistakes. It’s not about having correct judgment. It’s about avoiding incorrect judgments. [4]
COMPLEXITY THEORY
I was really into complexity theory back in the mid-90s. The more I got into it, the more I understand the limits of our knowledge and the limits of our prediction capability. Complexity has been super helpful to me. It has helped me come to a system that operates in the face of ignorance. I believe we are fundamentally ignorant and very, very bad at predicting the future. [4]
ECONOMICS
Microeconomics and game theory are fundamental. I don’t think you can be successful in business or even navigate most of our modern capitalist society without an extremely good understanding of supply-and-demand, labor-versus-capital, game theory, and those kinds of things. [4]
Ignore the noise. The market will decide.
PRINCIPAL-AGENT PROBLEM
To me, the principal-agent problem is the single most fundamental problem in microeconomics. If you do not understand the principal-agent problem, you will not know how to navigate your way through the world. It is important if you want to build a successful company or be successful in your dealings.
It’s a very simple concept. Julius Caesar famously said, “If you want it done, then go. And if not, then send.” What he meant was, if you want it done right, then you have to go yourself and do it. When you are the principal, then you are the owner—you care, and you will do a great job. When you are the agent and you are doing it on somebody else’s behalf, you can do a bad job. You just don’t care. You optimize for yourself rather than for the principal’s assets.
The smaller the company, the more everyone feels like a principal. The less you feel like an agent, the better the job you’re going to do. The more closely you can tie someone’s compensation to the exact value they’re creating, the more you turn them into a principal, and the less you turn them into an agent. [12]
I think at a core fundamental level, we understand this. We’re attracted to principals, and we all bond with principals, but the media and modern society spend a lot of time brainwashing you about needing an agent, an agent being important, and the agent being knowledgeable. [12]
COMPOUND INTEREST
Compound interest—most of you should know it in the finance context. If you don’t, crack open a microeconomics textbook. It’s worth reading a microeconomics textbook from start to finish.
An example of compound interest—let’s say you’re earning 10 percent a year on your $1. The first year, you make 10 percent, and you end up with $1.10. The next year, you end up with $1.21, and the next year $1.33. It keeps adding onto itself. If you’re compounding at 30 percent per year for thirty years, you don’t just end up with ten or twenty times your money—you end up with thousands of times your money. [10]
In the intellectual domain, compound interest rules. When you look at a business with one hundred users growing at a compound rate of 20 percent per month, it can very, very quickly stack up to having millions of users. Sometimes, even the founders of these companies are surprised by how large the business scales. [10]
BASIC MATH
I think basic mathematics is really underrated. If you’re going to make money, if you’re going to invest money, your basic math should be really good. You don’t need to learn geometry, trigonometry, calculus, or any of the complicated stuff if you’re just going into business. But you want arithmetic, probability, and statistics. Those are extremely important. Crack open a basic math book, and make sure you are really good at multiplying, dividing, compounding, probability, and statistics.
BLACK SWANS
There’s a new branch of probability statistics, which is really around tail events. Black swans are extreme probabilities. Again, I have to refer back to Nassim Taleb, who I think is one of the greatest philosopher-scientists of our times. He’s really done a lot of pioneering work on this.
CALCULUS
Calculus is useful to know, to understand the rates of change and how nature works. But it’s more important to understand the principles of calculus—where you’re measuring the change in small discrete or small continuous events. It’s not important you solve integrals or do derivations on demand, because you’re not going to need to in the business world.
FALSIFIABILITY
Least understood, but the most important principle for anyone claiming “science” on their side—falsifiability. If it doesn’t make falsifiable predictions, it’s not science. For you to believe something is true, it should have predictive power, and it must be falsifiable. [11]
I think macroeconomics, because it doesn’t make falsifiable predictions (which is the hallmark of science), has become corrupted. You never have a counterexample when studying the economy. You can never take the US economy and run two different experiments at the same time. [4]
IF YOU CAN’T DECIDE, THE ANSWER IS NO.
If I’m faced with a difficult choice, such as:
→ Should I marry this person?
→ Should I take this job?
→ Should I buy this house?
→ Should I move to this city?
→ Should I go into business with this person?
If you cannot decide, the answer is no. And the reason is, modern society is full of options. There are tons and tons of options. We live on a planet of seven billion people, and we are connected to everybody on the internet. There are hundreds of thousands of careers available to you. There are so many choices.
You’re biologically not built to realize how many choices there are. Historically, we’ve all evolved in tribes of 150 people. When someone comes along, they may be your only option for a partner.
When you choose something, you get locked in for a long time. Starting a business may take ten years. You start a relationship that will be five years or maybe more. You move to a city for ten to twenty years. These are very, very long-lived decisions. It’s very, very important we only say yes when we are pretty certain. You’re never going to be absolutely certain, but you’re going to be very certain.
If you find yourself creating a spreadsheet for a decision with a list of yes’s and no’s, pros and cons, checks and balances, why this is good or bad…forget it. If you cannot decide, the answer is no. [10]
RUN UPHILL
Simple heuristic: If you’re evenly split on a difficult decision, take the path more painful in the short term.
If you have two choices to make, and they’re relatively equal choices, take the path more difficult and more painful in the short term.
What’s actually going on is one of these paths requires short term pain. And the other path leads to pain further out in the future. And what your brain is doing through conflict avoidance is trying to push off the short-term pain.
By definition, if the two are even and one has short-term pain, that path has long-term gain associated. With the law of compound interest, long-term gain is what you want to go toward.
Your brain is overvaluing the side with the short-term happiness and trying to avoid the one with short-term pain.
So you have to cancel the tendency out (it’s a powerful subconscious tendency) by leaning into the pain. As you know, most of the gains in life come from suffering in the short term so you can get paid in the long term.
Working out for me is not fun; I suffer in the short term, I feel pain. But then in the long term, I’m better off because I have muscles or I’m healthier.
If I am reading a book and I’m getting confused, it is just like working out and the muscle getting sore or tired, except now my brain is being overwhelmed. In the long run I’m getting smarter because I’m absorbing new concepts from working at the limit or edge of my capability.
So you generally want to lean into things with short-term pain, but long-term gain.
What are the most efficient ways to build new mental models?
Read a lot—just read. [2]
Reading science, math, and philosophy one hour per day will likely put you at the upper echelon of human success within seven years.
收集心智模型
在决策过程中，大脑是一个记忆预测的机器。
一种糟糕的记忆预测方式是“过去发生了X，所以未来也会发生X。”这种方式过于依赖具体情境。你需要的是原则，你需要的是心智模型。
我找到的最好的心智模型来源于进化论、博弈论以及查理·芒格。查理·芒格是沃伦·巴菲特的合伙人，是一位非常出色的投资者。他有大量优秀的心智模型。作家兼交易员纳西姆·塔勒布也有很好的心智模型。本杰明·富兰克林同样如此。我基本上把自己的头脑装满了这些心智模型。[4]
我用自己的推文和其他人的推文作为格言，帮助压缩我的学习并回忆它们。大脑空间是有限的——你的神经元有限——所以你可以把这些模型看作是指针、地址，或助记符，帮助你记住那些深植于内心的原则，并且你有支持它们的基础经验。
如果你没有相关的基础经验，这些只是一些引语的集合。它们可能很酷，一时令人鼓舞，也许你会做出一个漂亮的海报。但之后你会忘记，然后继续前进。心智模型实际上是让你回忆起自己知识的简洁方式。[78]
进化
我认为很多现代社会的现象可以通过进化来解释。有一种理论是文明存在的目的是解决“谁有资格繁衍”的问题。从纯粹的性选择角度看，精子丰富而卵子稀缺，这实际上是一个分配问题。
几乎所有人类的成就都可以追溯到人们试图解决这个问题。
进化、热力学、信息论和复杂性在生活的很多方面具有解释和预测的力量。[11]
反向思维
我不相信自己有能力预测什么会成功。相反，我尝试排除那些不会成功的东西。我认为成功只在于不犯错误。这并不是关于做出正确的判断，而是避免错误的判断。[4]
复杂性理论
在90年代中期，我非常痴迷于复杂性理论。越深入了解它，我越明白我们知识的局限和预测能力的局限。复杂性对我非常有帮助，它帮助我构建了一个在无知情况下依然能运作的系统。我相信我们本质上是无知的，对于未来的预测非常非常糟糕。[4]
经济学
微观经济学和博弈论是基础。如果你不具备对供需关系、劳动与资本、博弈论等方面的深刻理解，我认为你很难在商业上成功，甚至难以在现代资本主义社会中找到方向。[4]
忽略噪音，市场会做出决定。
委托代理问题
在我看来，委托代理问题是微观经济学中最根本的问题。如果你不理解委托代理问题，你就无法在这个世界上游刃有余。想要建立一家成功的公司或在人际交往中取得成功，这一点非常重要。
这个概念非常简单。尤利乌斯·凯撒有句著名的话：“如果你想做好某件事，那就亲自去做。如果不想，那就派别人去做。”意思是，如果你想把事情做好，那就必须亲自去做。当你是委托人时，你是拥有者，你会在意，并且会做好。而当你是代理人，代表他人去做事时，你可能会做得不好，因为你不在乎。你会为了自己的利益而优化，而不是为了委托人的资产。
公司越小，每个人越觉得自己是委托人。越少有代理人的感觉，你的工作就会做得越好。你越能把某人的报酬与他们创造的价值直接挂钩，就越能让他们成为委托人，而不是代理人。[12]
从核心上看，我们对这一点有本能的理解。我们会被委托人吸引，并与他们建立联系，但媒体和现代社会花费大量时间洗脑我们，说我们需要代理人，代理人很重要，代理人是知识渊博的。[12]
复利
复利——大多数人在金融领域应该了解它。如果你不了解，可以翻开一本微观经济学教科书。从头到尾读一遍是值得的。
复利的一个例子——假设你每年赚取10%的收益，从1美元开始。第一年你赚了10%，得到了1.10美元。第二年你得到了1.21美元，第三年1.33美元。它会不断积累。如果你每年以30%的速度复利三十年，你不仅仅是得到了十倍或二十倍的钱——你将得到上千倍的财富。[10]
在智力领域，复利同样适用。当你看到一个拥有一百名用户的企业，每月以20%的复合增长率增长时，它很快就会累积到拥有数百万用户。有时，甚至这些公司的创始人也会对企业规模的增长感到惊讶。[10]
基础数学
我认为基础数学非常被低估。如果你要赚钱，或者投资金钱，你的基础数学能力应该非常好。如果你只是进入商业领域，你不需要学习几何、三角函数、微积分这些复杂的东西。但你需要掌握算术、概率和统计学。这些非常重要。翻开一本基础数学书，确保你对乘法、除法、复利、概率和统计学非常熟练。
黑天鹅事件
概率统计中有一个新的分支，主要研究尾部事件。黑天鹅事件是极端的概率事件。我必须再次提到纳西姆·塔勒布，我认为他是我们时代最伟大的哲学家兼科学家之一。他在这一领域做了大量开创性的工作。
微积分
微积分是有用的，可以帮助理解变化率以及自然如何运作。但更重要的是理解微积分的原理——你在测量小的离散事件或连续事件的变化。不需要你随时解积分或做导数运算，因为在商业世界中你不会用到这些。
可证伪性
最不被理解，但对于任何声称“科学”的人来说却是最重要的原则——可证伪性。如果它不能提出可证伪的预测，那就不是科学。要相信某件事是真的，它应该具有预测能力，并且必须是可证伪的。[11]
我认为宏观经济学因为无法提出可证伪的预测（这是科学的标志），已经变得腐化。在研究经济时，你永远不会有对照实验。你无法同时对美国经济做两个不同的实验。[4]
如果你无法决定，答案就是“不”
当我面临一个艰难的选择时，比如：
→ 我应该和这个人结婚吗？
→ 我应该接受这份工作吗？
→ 我应该买这栋房子吗？
→ 我应该搬到这个城市吗？
→ 我应该和这个人合伙做生意吗？
如果你无法决定，答案就是“不”。原因是，现代社会充满了各种选择。选项太多了。我们生活在一个有七十亿人口的星球上，并且通过互联网与所有人相连。成千上万的职业摆在你面前，有太多选择。
你在生物上并不具备意识到有这么多选择的能力。历史上，我们进化于150人的部落中。当某人出现时，他们可能是你唯一的伴侣选择。
当你做出某个选择时，你会被锁定很长一段时间。创业可能需要十年。你开始一段关系，可能持续五年甚至更久。你搬到一个城市，可能是十到二十年。这些都是非常、非常长久的决定。我们必须在非常确定时才说“是”。你永远不可能绝对确定，但你会非常确定。
如果你发现自己在为一个决策创建电子表格，列出各种优缺点、好坏平衡……忘了它吧。如果你无法决定，答案就是“不”。[10]
向上跑
简单的启发：如果你在一个艰难的决策中难以抉择，选择短期内更痛苦的道路。
如果你面临两个选择，并且它们是相对平等的选择，那么选择那个在短期内更困难和痛苦的道路。
实际上发生的情况是，其中一条道路需要短期的痛苦，而另一条道路的痛苦则出现在未来。你的大脑通过回避冲突试图推迟短期的痛苦。
按定义，如果两者相等而一条有短期痛苦，那么这条道路必然会带来长期的收益。根据复利定律，长期收益是你要追求的。
你的大脑高估了短期快乐的一面，试图避免短期的痛苦。
所以你需要抵消这种倾向（这是一种强大的潜意识倾向），通过迎接痛苦。你知道，生活中的大多数收获都来自于短期的痛苦，换取长期的回报。
对我来说，锻炼身体并不有趣，我在短期内承受痛苦，但从长远来看，我变得更健康、更强壮。
如果我在阅读一本书时感到困惑，这就像锻炼肌肉一样，肌肉会酸痛疲劳，区别只是现在我的大脑被压得透不过气。但从长远来看，我变得更聪明，因为我正在努力吸收新概念，突破能力的边界。
所以，你通常应该倾向于那些带来短期痛苦但长期收益的事物。
那么，建立新心智模型的最有效方法是什么？
大量阅读——只要读就行了。[2]
每天阅读一小时科学、数学和哲学，可能会在七年内让你处于人类成功的上层阶级。