TB: Poor Charlie's Almanack (Munger)

• You can't really know anything if facts don't hang together on a latticework of theory — they won't be in usable form.
• Must have multiple models; with just one or two, human psychology will cause you to torture reality to fit them.
• Models must come from multiple disciplines.
• Ultra-simple problem-solving notions: decide big no-brainer questions first; apply numerical fluency; invert; apply elementary multidisciplinary wisdom; watch out for lollapalooza effects (confluences of biases).

Key ones from the Psychology of Human Misjudgment:

• Reward/punishment super-response: never think about anything else when you should be thinking about incentives.
• Doubt-avoidance: triggered by puzzlement + stress — leads to premature commitment, especially in religion.
• Social-proof tendency: when unsure, we copy others — learn to ignore examples that are wrong.
• Availability-misweighing: vivid evidence is over-weighted; abstract evidence is under-weighted.
• Lollapalooza: extreme consequences arise from confluences of multiple biases acting in the same direction.

• Any year you don't destroy one of your best-loved ideas is probably a wasted year.
• The ethos of not fooling yourself is one of the best you can possibly have. It's powerful because it's so rare.
• Of all forms of useful pride, perhaps the most desirable is a justified pride in being trustworthy.
• I feel I'm not entitled to an opinion unless I can state the arguments against my position better than those in opposition.
• The area between everything you should do and everything you can do without legal trouble should be huge.

• The idea that it's hard to find good investments, so concentrate in a few, is obviously good — yet 90% of the investment world doesn't think this way.
• Never ask for someone's opinion or forecast; ask what they have in their portfolio.
• People calculate too much and think too little.