Herd behaviour, social media and outside opinions

What you will learn

• How herd behaviour and information cascades form in markets and online.
• Why impressive track records on social media can appear by pure chance (survivorship bias).
• What EU supervisors require from people who post investment recommendations.
• How to use an outside-opinion filter before an idea from others reaches your plan.

Explanation

Herd behaviour means following what many others do instead of your own analysis. It feels safe because being wrong together seems less painful than being wrong alone. In markets, herds can push prices far in one direction and then reverse sharply.

An information cascade occurs when people ignore their own information and copy earlier decisions. Each new follower makes the idea look more convincing, even though no new information was added. On social media, likes, shares and comments work as social proof: popularity is mistaken for correctness. Feeds are usually ranked by engagement, so dramatic, confident posts are shown more often than careful ones.

Survivorship bias means you see only the survivors. Accounts that made lucky calls stay visible; accounts that were wrong delete posts or disappear. An echo chamber is an environment where you mostly hear opinions that confirm your own, which strengthens the confirmation bias from lesson 3.

Scams, fake gurus and paid signal groups were covered in the beginner trading course. Here the focus is on honest but unhelpful influence: well-meant opinions that still pull you away from your plan.

What supervisors say

In February 2024 the European Securities and Markets Authority (ESMA) and national supervisors published a reminder that people posting investment recommendations on social media may fall under the EU Market Abuse Regulation (MAR). Among other things, such recommendations must be presented objectively and conflicts of interest must be disclosed. In Germany, BaFin is the competent supervisor. A missing disclosure is a warning sign; check the current rules if you need details.

Worked example

Hypothetical example: 1,000 accounts post "up" or "down" calls on an index each week. Assume each call is a pure coin flip (50% chance of being right) and calls are independent.

1. Chance of 5 correct calls in a row = 0.55 = 1 ÷ 32 ≈ 3.1%.
2. Expected number of accounts with a perfect 5-week record = 1,000 ÷ 32 ≈ 31 accounts, all with zero skill.
3. Chance of at least 8 correct out of 10 calls: (45 + 10 + 1) ÷ 1,024 = 56 ÷ 1,024 ≈ 5.5%. Expected number: 1,000 × 0.055 ≈ 55 accounts.

If you see only these winners in your feed, they look like experts. The 945 others are invisible. A short public track record is therefore weak evidence of skill, which links to lesson 8: small samples are dominated by variance.

The outside-opinion filter

Before any outside idea reaches your watchlist, answer five questions in writing:

1. Does it fit one of my written setups? If not, stop.
2. Can I see the full, verifiable track record, including losses?
3. Has the author disclosed positions or payments?
4. Would I take this trade if I had found it myself, at this price and size?
5. Am I feeling urgency or fear of missing out? If yes, apply the protocol from lesson 5.

Common mistake

Practice activity

On paper: (a) Assume 2,000 accounts make random weekly calls with a 50% hit rate. How many would you expect to have 6 correct calls in a row? (b) Take one trade idea you saw recently online and run it through the five-question filter. (c) List the accounts you follow and mark which disclose losses and positions.

Key takeaways

• Herds and information cascades make ideas look stronger without adding new information.
• Survivorship bias creates impressive-looking records from pure chance.
• EU rules require objective presentation and disclosure of conflicts for investment recommendations; missing disclosure is a red flag.
• Run every outside idea through a written filter that starts with your own setups.

Next step

In the final lesson you will combine everything into a personal psychology plan with a self-assessment and an improvement loop.