groupthink in stock market group chats

Groupthink in a group chat is not people being stupid — it is competent members each privately holding a reservation and each concluding, wrongly, that they are the only one. The doubts exist but are never pooled, so the group acts on a consensus that no individual member actually holds. The tell is the absence of a specific objection rather than the loudness of the agreement: nine positive messages and not one naming a way this could fail is nine people watching each other.

a group acting on a consensus that no individual member privately holds is called

Groupthink. It is a failure of information flow rather than of intelligence — private reservations go unstated, silence gets recorded as agreement, and whoever posts first sets a frame that every later message merely edits. The fixes are procedural rather than attitudinal: have each person write their view before anybody speaks, and give somebody the explicit job of arguing the other side so that holding that position costs nothing socially.

how many people should I share an investment thesis with

One, chosen deliberately: somebody who does not hold the stock, whose own view you have not heard, and who will ask for your reasons rather than your returns. The usefulness of an outside view falls as the audience grows, because a large group is a place to perform and a single reader is a place to think. Write the reasoning down before you open the thread — otherwise what you produce is an adjustment to the room’s view, and you cannot tell the difference from inside.

what does it mean to steelman the bear case

Steelmanning the bear case means stating the argument against a holding in its strongest form — well enough that somebody who genuinely holds that view would accept your version as fair. It is the test of whether a group has considered the downside or merely dismissed a weak imitation of it. If nobody in the room can put the case against convincingly, the case against has not been examined.

does announcing a stock call publicly make it harder to change your mind

Yes, and the effect switches on the moment the view is stated rather than when it is formed. New evidence stops being weighed on its own and starts being weighed against what a reversal would cost in front of people who remember — that is commitment bias. The cheap defence is to announce the exit condition alongside the view, naming what would have to happen for you to be wrong, so that changing course reads as a plan being followed rather than a public defeat.

Groupthink and stock tips in group chats

The engineering batch group has forty-one people in it and has been running since 2016. On a Wednesday somebody posts a chart of a small cap that has trebled, and within an hour there are eleven messages, nine of them agreeing. You have actually read this company’s last two annual reports and you have a genuine reservation about how the receivables have moved. You type it out, read it back, soften two sentences, and post it. Four people react. Two ask you what you would buy instead. And somewhere in the next thirty seconds a thing has changed that you will not notice for six months: you no longer hold a view about a company. You hold a position in front of people.

An earlier lesson in this track argued that individual investors suffer from having nobody to answer to, and recommended building accountability deliberately. That is right, and it is incomplete in a way that does real damage, because accountability is not one thing. Answering to the right audience in the right form makes reasoning measurably better. Answering to the wrong audience in the wrong form makes it worse, and the wrong version is the one that is free, instant and already installed on your phone.

Think of it like this

The rehearsal and the show

A musician working alone will stop mid-phrase, go back, try it four ways and keep the ugly one that is closer to right. Put an audience in the room and the same musician plays through the mistake without stopping, because stopping is now a visible failure. The playing has not got worse. The willingness to be seen correcting has gone, and correcting was where the improvement was happening.

In the market

That is precisely what an audience does to an investment view. Nothing about your analysis degrades. What goes is the freedom to say “I have changed my mind” cheaply — and since almost all of the value in a long investing life comes from changing your mind at a reasonable cost, that is a very expensive thing to give away to a group chat.

Three separate effects, which people run together

The effect	When it operates	What it looks like from inside
The room’s view arrives first	Before you have formed one	You believe you reasoned your way to a conclusion. What actually happened is that you started from theirs and adjusted. Social proof does not feel like being persuaded; it feels like the question having an obvious answer
The correction becomes public	After you have stated a view	New evidence stops being weighed on its own. It gets weighed against what reversing would cost in front of people who remember. This is commitment bias, and having said it out loud is what switches it on
Conventional failure looks safer	While choosing between options	You quietly prefer the choice that, if it goes wrong, goes wrong the way everybody else’s went wrong. Losing money in the index nobody asks you about. Losing it in something you argued for alone, they do

The third is the least discussed and the most costly over a career, because it never produces a single identifiable mistake — only a lifetime of slightly worse choices, each of which was defensible.

Two audiences, one of which improves your reasoning

Makes decisions better

Does not already know your conclusion, so you have to state the case rather than defend it
Whose own view you have not heard, so you cannot anchor on it
Expects reasons, and asks about the process rather than the profit
Small — one or two people — so “I do not know” costs nothing to say
Has no position in the thing, and nothing riding on being right in front of the same crowd
Sees the decisions that did not work as well as the ones that did

Makes decisions worse

Has a visible consensus that arrives before you have thought
Scores you on outcome — a screenshot of a gain is applause and a loss is silence
Large enough that reversing requires a public correction
Full of people who hold the same position and need it to work
Rewards confidence and treats hedged language as weakness
Keeps no record, so only the calls that worked are ever mentioned again

Groupthink is not people being stupid

Groupthink is usually described as a group converging on a bad answer, which makes it sound like a failure of intelligence. It is not. It is a failure of information flow, and it happens between competent people who each privately hold a reservation and each conclude, correctly, that they are the only one. The reservations exist. They are never pooled. The group then acts on a consensus that no individual member actually holds, and every member walks away believing everybody else was confident.

The tell is the absence of a specific objection, not the presence of agreement. Nine messages saying the same positive thing and not one of them naming a way this could fail is not nine people agreeing. It is nine people watching each other.
Order of speaking decides the outcome. Whoever posts first sets the frame, and every subsequent message is an edit of it. In a group where one person is senior, older or wealthier, their view arriving first ends the discussion before it starts.
Silence is recorded as agreement, and almost never is. This is worth stating flatly because it governs family conversations about money in most Indian households, where disagreeing with an elder in front of others is simply not done.
The fix is procedural. Ask each person to write their view before anybody speaks, or specifically ask the quietest person for the case against. Assigning somebody the job of arguing the other side removes the social cost of holding that position, which is what was suppressing it.
Steelmanning is the test. If nobody in the group can state the bearish case well enough that a bear would accept it, the group has not considered it — it has only dismissed a weak version of it.

Building the version that helps

Four adjustments, none of which cost anything

1
Write before you read
The view, in five lines, with the reason and what would falsify it, saved with a date — before opening the thread, the forum or the video. If you then find you agree with the group, you have genuinely agreed with them. If you never write it first, you will never again know which of your views are yours.
2
Choose one person, not forty
One person who does not hold the stock, has not told you what they think, and is willing to ask what would have to be true for you to be wrong. That is the entire apparatus. Its usefulness is inversely proportional to the size of the audience, because a group is a place to perform and a single reader is a place to think.
3
Report process, not P&L
What you tell your one person is what you decided and why, at the time you decided it. Not what it did afterwards. An audience that scores you on outcomes will teach you to take decisions that look good, and an audience that asks for your reasoning will teach you to have some.
4
Make reversing cheap in advance
State the exit condition at the same time as the view, in public if the view was public: “I own this because of X; if the receivables do not normalise by the March quarter I will be wrong and I will say so.” A reversal you announced in advance is a plan being followed rather than a public defeat, and that one sentence is what buys back your freedom to change your mind.

◆ Your call

The auditor resigns in June

In January you posted a careful, well-argued positive note on a mid-sized company to the batch group of forty-one, and several people bought it. In June the statutory auditor resigns and the reason given in the filing is not a reassuring one. You now think you were wrong.

Check yourself

You want an outside view on a holding you are considering. Which arrangement is most likely to improve the decision rather than degrade it?

Simple bhasha mein

Batch group mein bol diya, ab peeche kaise hatein

Ektalis logon ka group, gyarah message pehle se haan-haan kar rahe hain — aur aap sochte ho ki maine khud socha. Nahi. Aapne unke jawab mein thoda sudhaar kiya, bas. Aur jis din aapne apni raay post kar di, us din se koi bhi nayi khabar company ke hisaab se nahi tuloti — "ab palatna padega, sab dekhenge" ke hisaab se tulti hai. Ilaaj mehnat ka nahi, tarteeb ka hai: thread kholne se pehle paanch line likh lo — kyun le rahe ho, aur kya nikle toh galat maanoge. Aur raay ke saath hi exit ki shart bhi bol do — tab palatna haar nahi, plan ban jaata hai.

What to remember

Accountability is not uniformly good: the audience and the format decide whether it improves reasoning or damages it.
Hearing the room before you write means your view is an adjustment to theirs, and you cannot tell from inside.
Once a view is stated publicly, new evidence gets weighed against the cost of a public correction instead of on its own.
Groupthink is competent people each privately doubting and each assuming they are alone; the tell is no specific objection, not loud agreement.
Announce the exit condition with the view — a reversal you planned for is a plan, not a defeat.

You reached the endMark it done and keep your streak going.

Up nextRunning somebody else’s money Previous: The loss nobody at home knows about

Finished this lesson?

Mark it done to track your progress through the curriculum.

Common questions

Short, direct answers to what people ask about this topic.

groupthink in stock market group chats: Groupthink in a group chat is not people being stupid — it is competent members each privately holding a reservation and each concluding, wrongly, that they are the only one. The doubts exist but are never pooled, so the group acts on a consensus that no individual member actually holds. The tell is the absence of a specific objection rather than the loudness of the agreement: nine positive messages and not one naming a way this could fail is nine people watching each other.
a group acting on a consensus that no individual member privately holds is called: Groupthink. It is a failure of information flow rather than of intelligence — private reservations go unstated, silence gets recorded as agreement, and whoever posts first sets a frame that every later message merely edits. The fixes are procedural rather than attitudinal: have each person write their view before anybody speaks, and give somebody the explicit job of arguing the other side so that holding that position costs nothing socially.
how many people should I share an investment thesis with: One, chosen deliberately: somebody who does not hold the stock, whose own view you have not heard, and who will ask for your reasons rather than your returns. The usefulness of an outside view falls as the audience grows, because a large group is a place to perform and a single reader is a place to think. Write the reasoning down before you open the thread — otherwise what you produce is an adjustment to the room’s view, and you cannot tell the difference from inside.
what does it mean to steelman the bear case: Steelmanning the bear case means stating the argument against a holding in its strongest form — well enough that somebody who genuinely holds that view would accept your version as fair. It is the test of whether a group has considered the downside or merely dismissed a weak imitation of it. If nobody in the room can put the case against convincingly, the case against has not been examined.
does announcing a stock call publicly make it harder to change your mind: Yes, and the effect switches on the moment the view is stated rather than when it is formed. New evidence stops being weighed on its own and starts being weighed against what a reversal would cost in front of people who remember — that is commitment bias. The cheap defence is to announce the exit condition alongside the view, naming what would have to happen for you to be wrong, so that changing course reads as a plan being followed rather than a public defeat.