I thought I was a good decision maker.
At work, the decisions came quickly and they usually held up. Strategy, restructuring, hiring, negotiation — I trusted my judgement and my track record gave me reasons to. Then I started trading seriously. And then I looked more carefully at my personal relationships. And I discovered something that took a long time to fully accept.
My decisions were not as good as I thought. And the errors were not random — they followed patterns. The same patterns, appearing in the trading account, in the career choices, in the relationships, dressed in different clothes but running on the same underlying software.
The most painful discovery was not about markets or money. It was about people. I had been assuming, unconsciously and consistently, that the people around me thought roughly the way I did. That they ran similar calculations, weighted similar values, responded to similar signals. They do not. Even the closest ones have completely different equations running. What feels obvious to you is invisible to them. What feels obvious to them never occurred to you. Most of the disappointments in my relationships — and there have been some significant ones — trace back to this assumption more than anything else.
The word for this pattern is cognitive bias. There are 59 documented ones. I know because I have the list. What I want to share here are the six I keep recognising in myself — in the trading account, in the career, in the relationships — because seeing them clearly has been the only thing that has made any difference.
· · ·
|
"We are blind to our blindness. We have very little idea of how little we know." — Daniel Kahneman, Thinking, Fast and Slow |
· · ·
Anchoring. A few years ago I was looking at real estate. I had a number in my head — what the property should cost based on prices I had seen before. The market had moved. The numbers I was comparing against were outdated. But my brain kept returning to the old reference point, treating it as the correct baseline against which the current price was being evaluated. Too expensive, I decided. I did not buy. Since then the price has roughly doubled. The anchor was wrong. The decision followed the anchor anyway.
Anchoring is what happens when the first number you encounter becomes the invisible reference point for every number that follows. In trading it is the price you paid for a position — you cannot think clearly about whether to sell because your brain keeps comparing the current price to your entry price rather than asking whether it is a good price today. In relationships it is the first version of a person you met — you keep comparing current behaviour to that original impression rather than seeing what is actually in front of you now.
Confirmation bias. Imagine you buy a stock because you believe the sector is undervalued. After the purchase, you read every article that confirms the sector is undervalued. The articles that argue the opposite — you find them less credible, less rigorous, written by people who do not understand the full picture. You are not being dishonest. You are doing exactly what the human brain does automatically. It looks for evidence that supports the existing belief and discounts evidence that challenges it. The position goes against you. You find more reasons to hold. The losses grow.
In relationships, confirmation bias is the reason that once you have decided someone is trustworthy, you interpret ambiguous behaviour as trustworthy. And once you have decided someone is not, you interpret neutral behaviour as further evidence of the problem. The decision precedes the evidence. The evidence is then selected to match the decision.
Sunk cost. I spent the majority of my career in telecommunications. By the time it became clear to me that the sector was structurally declining — consolidating, commoditising, losing the strategic importance it once had — I had invested so much. Two decades of relationships, expertise, reputation, identity. The rational question was whether the future of the sector justified continuing to invest in it. The question I was actually asking was whether walking away would mean those twenty years were wasted. They are different questions. I stayed longer than the rational analysis would have supported. The time already spent should not have been a factor in the decision about what to do next. It was the primary factor.
Sunk cost thinking keeps people in bad trades, bad jobs, bad relationships. Not because the future looks good. Because the past feels like it needs to be justified.
Loss aversion. Kahneman's research showed that losses feel approximately twice as painful as equivalent gains feel good. Losing one hundred euros hurts roughly twice as much as gaining one hundred euros feels positive. This asymmetry distorts almost every financial decision people make — and most relationship decisions too.
In trading: you sell winners quickly to lock in the good feeling of a gain, and you hold losers far too long to avoid realising the pain of a loss. The result is a portfolio full of losing positions and no winning ones. In relationships: you stay in situations that are not working because the pain of the loss — of the person, the history, the identity you built around the relationship — feels larger than any potential gain from leaving. Loss aversion is not irrational. It is deeply human. It is also frequently wrong about the actual arithmetic.
Halo effect. I had a friend. Good company, generous socially, someone I genuinely enjoyed. We had years of good history together — dinners, conversations, the kind of easy familiarity that feels like it means something. At some point I needed a professional reference. Something specific, a genuine recommendation that required a small effort. He ignored the request completely. Not a no. Not an explanation. Just silence.
The halo from the social relationship had extended, in my mind, into professional reliability. Into genuine care for my wellbeing beyond the pleasant evenings. The halo was not accurate. The person who was excellent company was also someone who would not inconvenience himself when it mattered. Both things were true. I had only registered one of them.
The halo effect is the tendency to let one positive quality — charm, intelligence, success, attractiveness — cast a glow over all other qualities, making us assume the whole person matches the part we can most easily see. It explains why we trust certain people far beyond what the evidence supports. And why we are surprised when they turn out to be more complicated than the halo suggested.
Recency bias. When markets keep rising, it starts to feel like they will keep rising. Not as a considered forecast but as a feeling — a background assumption that what has been happening will keep happening. The same thing happens in careers. When everything is going well — the role is solid, the income is growing, the position feels secure — it starts to feel permanent. Not because you have analysed the structural trends. Because recent experience has been good and the brain extrapolates from recent experience without sufficient adjustment for longer-term patterns.
I watched this happen in the telecommunications sector across an entire career. The companies that had been growing for twenty years assumed they would keep growing. The executives who had been rising assumed they would keep rising. The sector that had been essential assumed it would remain essential. The layoffs, when they came for people I knew, were not genuinely surprising from the outside. They were genuinely shocking to the people inside. Recency bias had made the recent past feel like a reliable forecast of the future.
· · ·
The honest conclusion is not that you can eliminate these biases. You cannot. They are not errors in an otherwise rational system. They are features of the system — shortcuts that worked well enough across most of human history to be deeply embedded. The brain that avoids losses aggressively survived better than the brain that did not. The brain that extrapolates from recent experience is usually right enough. The halo effect saves cognitive energy. None of these patterns are bugs.
What you can do is recognise them in the moment — or better, before the moment. The pre-mortem: before making a significant decision, ask what would have to be true for this to go badly wrong. You are deliberately looking for the disconfirming evidence, the alternative explanation, the scenario where your confident current view turns out to be anchored on the wrong reference point. It does not guarantee better decisions. It slightly improves the odds.
The deeper work is humility about the machinery. Most of what feels like judgement is pattern repetition. Most of what feels like insight is confirmation. Most of what feels like loyalty is sunk cost. Knowing this does not make you immune. It makes you slightly less surprised when you catch yourself doing it again — which you will, because so will I.
|
This week's insight You are not making decisions. You are running software. The anchors, the confirmation loops, the sunk costs, the loss aversion, the halos, the recency extrapolations — they are all running underneath what feels like clear thinking. You cannot uninstall them. But you can learn to notice them mid-process, which is the only upgrade available. We are blind to our blindness. The first step is admitting the blindness exists. |
If this landed, there is one piece every week at stillnavigating.com — honest writing about the stuck middle, from someone still in it.