Herd Mentality: Why You Followed the Crowd Into (or Out of) a Losing Position
Herd mentality is why you bought when everyone else was buying and sold when everyone else was panicking. The research behind crowd-driven bubbles and capitulations, plus a checklist to catch it before your next trade.
You didn’t do independent research on that token before you bought it. You saw the chart moving, the timeline filling up with people posting gains, and a feeling that everyone already knew something you didn’t — so you bought too. Later, when the same asset dropped fast and every feed you followed turned into a wall of panic, you sold into the worst of it, not because your thesis had changed but because everyone around you was selling. Both of those moments have the same cause: herd mentality, the tendency to copy the crowd’s behavior instead of running your own analysis, and it’s just as capable of wrecking an exit as it is an entry.
What herd behavior actually is
Herd behavior is well studied outside of finance — flocks, crowds, information cascades — and its application to markets has its own research history. The clearest starting point is Sushil Bikhchandani and Sunil Sharma’s review “Herd Behavior in Financial Markets,” published in IMF Staff Papers (Vol. 47, No. 3, 2001). Their core finding: herding can emerge purely from an information cascade — once enough people have visibly taken the same action, later participants rationally start ignoring their own private information and just copy what they see, because the crowd’s visible behavior seems to carry more signal than whatever they individually know. The catch is that once the cascade starts, no new private information ever makes it into the public pool — everyone’s just watching everyone else, and the “wisdom of the crowd” stops actually aggregating anything.
Three separate engines feed into that cascade, and it’s worth being able to name each one:
Informational herding. You copy the crowd because you assume they collectively know something you don’t. This is the mechanism Bikhchandani and Sharma modeled — individually rational, collectively fragile, because it only takes a handful of early movers (who might themselves be wrong, or trading on something irrelevant to you) to set the whole cascade off.
Social and emotional contagion. Watching other people’s fear or excitement in real time — a group chat, a comments section, a trending chart — activates the same feeling in you, independent of any information content at all. This is the fastest of the three, and it’s the one most amplified by always-on social feeds.
Reputational herding. Professional money managers sometimes mimic peers specifically to avoid being the one who was visibly wrong alone, since being wrong in a crowd is less career-damaging than being wrong solo. Retail traders have a version of this too: nobody wants to be the person who called the top wrong on a timeline full of people who called it right.
A real, documented example: GameStop, January 2021
The clearest recent case study of herding moving a real price, at scale, with a regulator’s own analysis attached, is the meme-stock episode of January 2021. In its Staff Report on Equity and Options Market Structure Conditions in Early 2021, the SEC concluded that GameStop’s share price kept climbing for weeks not primarily because short sellers were being forced to cover their positions — the popular “short squeeze” story at the time — but because of sustained, self-reinforcing bullish sentiment among individual investors, much of it coordinated and amplified through social media. The report put it directly: it was the positive sentiment, not the buying-to-cover, that sustained the weeks-long price appreciation.
That’s herding working exactly as the theory predicts: visible crowd behavior (a stock everyone was talking about, moving fast) became the reason to buy, for a large enough number of people that the reason became self-fulfilling for a while — right up until it wasn’t, and the people who bought closest to the top were left holding a position with no thesis except “it was moving.”
Why crypto herds harder
Herding isn’t unique to crypto, but several features of crypto markets make it stronger and faster there. Academic researchers studying crypto herding typically use a cross-sectional absolute deviation (CSAD) method, which detects herding by checking whether individual assets’ returns cluster unusually tightly around the market average during big moves — tighter clustering than normal market co-movement would predict is the signature of a crowd all doing the same thing at once rather than reacting to their own analysis. Multiple studies using this approach — including work published in the Journal of Risk and Financial Management examining liquidity and sentiment effects, and a 2023 study covering 77 cryptocurrencies from 2018 to 2023 — find that the herding signal tends to show up more strongly during periods of lower liquidity and sharper drawdowns, exactly the conditions a thinly traded token or a struggling exchange creates.
That tracks with what the asset class looks like structurally: crypto trades continuously, it’s heavily retail, its major information channels are social platforms rather than filed disclosures, and a large share of its market cap sits in assets thin enough that a relatively small wave of coordinated buying or selling moves price fast and visibly — which is exactly the kind of visible movement that triggers the next round of informational and social herding.
The two directions herding pulls you
Herding gets discussed almost entirely as a buying phenomenon — chasing pumps, buying tops — but the same mechanism runs in reverse during a crash, and the selling version is arguably more damaging to a recovery because it turns a paper loss into a realized one at the worst possible moment.
| Buying-herd (bubble entry) | Selling-herd (panic capitulation) | |
|---|---|---|
| Trigger | A visibly pumping chart, a trending token, a feed full of gains | A sharp drawdown, liquidation cascades, a feed full of panic |
| Dominant emotion | Anticipated regret at missing a gain (overlaps with FOMO) | Contagious fear of a loss getting worse |
| What it feels like from inside | “Everyone’s making money and I’m not” | “Everyone’s getting out, I need to too” |
| What it actually does | Buys in near the top, with no independent thesis | Sells near the bottom, converting an unrealized loss into a realized one |
| How it usually resolves | The rally runs out of new buyers and reverses | The selling exhausts itself and often rebounds without the herd |
| Related post on this site | FOMO Is Why You Bought the Top | Dead Cat Bounce or Real Recovery? |
Both columns share the same root cause — outsourcing the decision to the crowd’s visible behavior instead of your own analysis — which is why the fix for both is the same discipline, applied at different moments.
The checklist: are you herding right now?
Before you buy something because it’s moving, or sell something because everyone around you is panicking, run through this:
- Would I take this action if I hadn’t seen anyone else’s behavior first? If the honest answer is no — if the trigger is specifically that other people are doing it — that’s the herding tell, in either direction.
- Can I state my reasoning without referencing the crowd? Try writing the thesis for buying or selling in one sentence without using words like “everyone,” “trending,” “crashing,” or “it’s happening.” If the sentence falls apart without those words, there’s no independent thesis underneath it.
- Is the information I’m reacting to actually new, or is it just other people’s reaction to the same old information? A genuine new data point (an exchange insolvency, a protocol exploit, a regulatory ruling) is worth reacting to. A feed full of people reacting to that same data point a second time isn’t new information — it’s the cascade repeating itself.
- Am I moving faster than I normally would, specifically because I feel behind the crowd? Herding compresses decision time the same way FOMO does. A rushed decision made to keep pace with a group is a worse decision than the same one made on your own schedule.
Two or more “yes, I’m herding” answers is enough signal to pause before acting — the same threshold used for spotting revenge trading or a FOMO entry, because the underlying mechanism (skipping your own process in favor of a shortcut) is the same failure wearing different triggers.
What to do instead
Separate “the crowd is moving” from “I have a reason.” Both can be true at once, but only the second one should drive a trade. If you strip out every reference to what other people are doing and you’re still left with a reason, act on the reason. If you’re not, you were herding.
Build a pre-commitment rule for both directions. Most people only guard against buying tops. A rule like “I don’t sell a position during a liquidation cascade without a 24-hour cooling-off period, unless my original thesis has genuinely broken” protects against the panic-selling version of the same bias, which is the one that actually locks in losses.
Treat a trending or crashing asset as a prompt to slow down, not speed up. The more visible and urgent a crowd’s behavior feels, the more that urgency itself should be a signal to check your own reasoning before acting on theirs.
Remember that the crowd not being wrong yet isn’t the same as the crowd being right. A cascade can run for a long time before it corrects — that’s exactly why waiting for the price to prove you wrong is not a substitute for having a thesis you could defend before the move happened.
FAQ
Is herd mentality the same thing as FOMO? They’re linked but not identical. Herd mentality is the crowd’s behavior — a large number of people moving the same direction at the same time, buying or selling because everyone else is. FOMO is what that looks like inside one person watching the crowd: a specific fear of being left out of a gain. Herding can also pull you the other way, into panic selling during a crash, where the driving emotion isn’t FOMO at all — it’s contagious fear.
Is following the crowd ever the right move? Sometimes the crowd is right — a price move can reflect real, distributed information that no single trader has on their own, and ignoring it out of stubbornness isn’t automatically smarter. The problem isn’t that crowds are always wrong; it’s that herding specifically means you stopped checking whether they’re right this time, and started trading on the fact that they’re moving rather than on why. If you can articulate the why independent of the crowd, you’re not herding even if you end up doing the same thing as everyone else.
How do I tell herding apart from doing real analysis that happens to agree with the crowd? Ask whether your conclusion would survive if you’d seen the same information privately, with no visible crowd attached to it — no trending chart, no timeline full of people posting gains or panic. If your conviction depends on watching other people act first, that’s herding. If it would hold up on the data alone, it’s analysis that happens to overlap with what the crowd is doing, which is a different thing entirely.
Does herding happen more in crypto than in stocks? Academic studies using cross-sectional deviation methods generally find stronger herding signals in crypto markets than in traditional equities, and several studies report the effect intensifying during periods of low liquidity or sharp drawdowns — exactly the conditions a thinly traded token or an exchange in trouble tends to create. That doesn’t mean stock markets are immune, as the 2021 meme-stock episode showed; it means crypto’s smaller, more retail-heavy, more social-media-driven markets tend to make the crowd’s influence on price larger and faster.
If a herd-driven entry is the position you’re sitting on right now, Down 50%? Here’s the Actual Plan, Not the Pep Talk walks through deciding what to do with it on its current merits — independent of how it got opened.