Thinking, Fast and Slow — Kahneman: COMPLETE

System 1 is what makes you see anger in a face, hear irony in a voice, solve 2+2 and read a road sign at a distance — without you deciding to. System 2 is what computes 17×24 in the head, fills in the tax return and compares two valuation cases line by line. The difference is not intelligence but cost: System 2 has a limited budget of attention and prefers not to lay out more than necessary. Kahneman and his colleague Jackson Beatty measured effort in the pupils in the 1960s — in the Add-1 task (converting four digits, one at a time, up by one) the pupils dilated several percent, and in the harder Add-3 they stayed maximally dilated as long as the subjects endured.

The pupil is, as Kahneman writes, a more sensitive measure of mental strain than anything the subjects themselves reported.

The laziness of System 2 shows best in the bat-and-ball problem: a bat and a ball together cost 1,10 dollars, and the bat costs 1 dollar more than the ball — what does the ball cost? The intuitive answer, 10 cents, jumps out immediately and feels right. It is also wrong — the right answer is 5 cents — and in the original study more than half of the students at Harvard, MIT and Princeton answered wrong, at other universities over 80 percent.

The point is not that people are stupid but that System 2 approves an answer that feels coherent without running the check computation. Investors do the same thing every time a low P/E is automatically translated into »cheap« without the question cheap relative to what — the feeling of plausibility replaces the computation.

The Müller-Lyer illusion — two equally long lines where the arrowheads make one look longer — is Kahneman's closing image, and it is more uncomfortable than it looks: you can measure the lines, know that they are equal, and still see one as longer. Insight does not cure the illusion. That is why Kahneman's practical advice is never »try to see right« but »learn to recognize situations where you can be fooled, and use tools«.

In an investment context that means yardsticks outside the head: checklist, written thesis, base rates — not willpower. The material of the course in AKM1's bias track and the logging in /profil are in this light not bureaucracy but Müller-Lyer rulers: instruments that see right for you when the eye cannot.

The book's perhaps most underestimated concept is introduced here and returns in the chapter after chapter: WYSIATI — what you see is all there is. System 1 builds a coherent story from the information that is present and never asks the question what is missing. Three quarters in a row with numbers above expectation, a convincing CEO and a pretty product roadmap are enough for an entire investment case to feel complete — not because it is, but because the feeling of coherence is created within the frame of what is already on the screen.

The question »what information would change my picture, and have I had time to look for it?« is one of the cheapest System 2 interventions there are.

In 1973 Tversky and Kahneman had participants judge whether the letter K is more common as the first or third letter in English words. The majority answered first — because it is easier to recall words that begin with K than words where K sits in third place. In reality K is more common as the third letter.

The ease of memory was silently exchanged for frequency in the world. The mechanism is banal and the consequences enormous: what happened recently, what was dramatic, what you personally experienced weighs in on every unconscious risk assessment — regardless of true frequency.

The price tag of availability can be measured in human lives. After 11 september 2001 many Americans chose the car instead of flying; since car travel is riskier per mile, several hundred people died extra in road traffic in the months after the attack according to Gigerenzer's estimate — a hidden cost no news report carried, driven by images impossible to shake off. Kahneman writes that the world in your head is not an exact copy of reality: media deliver a sample produced by drama, not by frequency.

For the investor this means that the latest macro shock, the latest rally and the latest CEO scandal are overrepresented in every unwritten risk assessment — and that boring, slow risks like fees and inflation, which never get headlines, are systematically underestimated.

Kahneman confesses in the book his own pitfall as a young military psychologist: he taught that small samples are unreliable — and at the same time systematically chose microscopic samples in his own studies. He calls the belief that small samples mirror the population the law of small numbers, a parody of the law of large numbers. The classic example: a large American foundation noted that the most successful schools in the state's test data were unusually small, and bet billions of dollars on small schools.

But the worst schools were also unusually small — small samples produce extreme averages, at both ends. The investor's version: three strong quarters in a row for a growth company (V01) and the trend feels proven. It is not proven — it is a K-problem: easy to remember, hard to generalize from.

The antidote to both traps is questions, not rest. Against availability: how often does this actually happen per year, per decade, in all history — not when did it happen last? Against small numbers: how many observations does the conclusion build on, and what does the spread look like? WYSIATI from chapter 1 explains why the antidote must be a routine: System 1 will never on its own miss the information it has never seen.

A written requirement of five to ten years of data series before a growth assessment gets its own V number is precisely such a routine — it forces out what the news flow never delivers.