——
Level — · Streak —
— XP
Mastery—weighted
Quizzes 30d—last 30 days
Cert progress—active level
Rank—global
Loading...
The page behind this dialog is live. Create a free account or sign in and you'll land right back on it.
Why crowds are wrong at extremes and expertise is overestimated
In late January 2021, GameStop traded at \$483 a share intraday — up roughly 15x in three weeks on the back of a coordinated retail-driven short squeeze. By mid-February it had fallen to \$40. Every retail investor who bought at \$300 in 'because everyone was talking about it' lost roughly 87% in three weeks. Two years earlier, in February-March 2020, the S&P 500 fell 34% in 33 days as retail investors panic-sold their accounts. By August 2020, the index had fully recovered. The investors who held returned to even; the ones who sold near the lows locked in losses they could never recover. Both episodes are the same bias in different directions: the human instinct to do what the crowd is doing, accelerated by the velocity of modern markets and amplified by social media. Following the herd is the most expensive consistent mistake in retail investing.
is the universal investor instinct to do what the crowd is doing. Humans evolved as social animals; following the group was survival on the savanna. In markets, the same wiring creates bubbles (everyone buys what's going up; the buying pushes it higher; the higher price attracts more buyers) and panics (everyone sells what's going down; the selling pushes it lower; the lower price triggers more selling). Sentiment indicators — the AAII Investor Sentiment Survey, the CNN Fear & Greed Index, equity inflow/outflow data — repeatedly hit extreme readings at exactly the wrong times. At the March 2000 NASDAQ peak (5048, just before the dot-com crash), bullish sentiment was near record highs and retail inflows into tech funds were at all-time records. At the March 2009 bottom (S&P 500 at 676, the trough of the financial crisis), bearish sentiment hit record extremes and retail money was flooding out of equity funds. Both readings were essentially the loudest possible contrarian signal — and both were ignored by the people who could have benefited most.
is the emotional engine that turns mild herding into bubble dynamics. The mechanic is well-documented in social psychology: when everyone in your peer network is reporting gains on a particular asset (cryptocurrency, GameStop, Tesla in 2020-2021), the social pain of being the only person not participating becomes harder to bear than the financial risk of buying late. The 2021 GameStop episode is the canonical retail example: the stock went from \$20 in early January to \$483 in three weeks, driven primarily by retail buyers organizing on Reddit's r/wallstreetbets. Most retail buyers entered between \$200 and \$400 — fully aware the fundamentals didn't support the price, justifying the purchase as 'I'll just take a small position to participate.' Within four weeks the stock had fallen to \$40. The pattern repeats with every speculative bubble — pre-IPO cannabis stocks 2018-2019, NFTs 2021-2022, AI-themed names 2023-2024. The inflexion isn't whether the underlying technology is real (it often is); it's whether the buying decision is driven by independent valuation analysis or by the social pain of watching others profit.
is the bias that turns the herding instinct into a portfolio-destroying combination. The Dunning-Kruger effect describes the empirical relationship between competence and self-assessed competence: at low levels of skill, people systematically overestimate their ability (the 'peak of Mount Stupid' in the popular framing); at higher levels of expertise, people often UNDERestimate their ability because they've learned to see the gaps in their knowledge. Beginners are uniquely vulnerable to overconfidence in investing because the feedback loop is slow and noisy: a few successful trades in a bull market feel like skill, even when they're entirely the result of riding a rising tide. The disciplined check is to attribute returns to their actual sources: was that 30% gain because the market was up 25% (beta), because the sector was up 40% (sector rotation), or because the specific name outperformed both (alpha)? Most retail outperformance dissolves once the first two are subtracted; the genuine alpha component is much smaller than people think.
The empirical record on overconfidence in retail investing is stark. Brad Barber and Terrance Odean's 2000 paper 'Trading Is Hazardous to Your Wealth' (Journal of Finance) studied 66,465 retail brokerage accounts from 1991-1996. The most active traders (those in the top quintile by turnover) underperformed buy-and-hold by approximately 6.5 percentage points per year — a multi-decade-compounding gap that turns into orders-of-magnitude wealth differences. The mechanism is straightforward: every trade has friction costs (commissions in 1991-1996, today bid-ask spreads and tax drag), and most discretionary trading decisions add zero or negative information value. Combine 6.5 pp/yr underperformance with the math of compounding from m0_l1 and the conclusion is unambiguous: for retail investors, doing less is almost always better than doing more. The platform's tools are designed to be used quarterly per holding, not daily. Refresh frequency does not improve information quality.
Barber and Odean 2000 sorted 66,465 retail accounts by annualized turnover (the proportion of the portfolio traded each year) and computed risk-adjusted net returns by quintile. Findings: the average household earned approximately 16.4% gross return / 15.3% net return; the lowest-turnover quintile (~2.4% annual turnover, essentially buy-and-hold) earned ~17.5% net; the highest-turnover quintile (~258% annual turnover, multi-times-per-year portfolio replacement) earned only ~10.0% net. Net underperformance gap: ~7.5 pp/yr at the extremes; ~6.5 pp/yr for the top-quintile vs the average. The mechanism: turnover-related costs (commissions, bid-ask spreads, taxes) plus the signal-to-noise ratio of high-frequency discretionary decisions both work against the active trader. Barber and Odean's follow-up papers (2001 'Boys Will Be Boys' on gender; 2008 day-trader studies) consistently replicate the pattern. The practical implication: there is no documented retail population for which higher trading frequency improves risk-adjusted returns, and the gap widens with turnover. Source: Barber & Odean, 'Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,' Journal of Finance 55(2), April 2000.
Brad Barber (UC Davis) and Terrance Odean (UC Berkeley) obtained six years of trading records from a major U.S. discount broker covering 66,465 retail households (1991-1996). They sorted households by annualized turnover and computed risk-adjusted returns by quintile. The lowest-turnover quintile (essentially buy-and-hold investors with ~2.4% annual turnover) earned ~17.5% net annualized returns. The highest-turnover quintile (~258% annual turnover — meaning the portfolio was replaced 2-3 times per year on average) earned ~10.0% net. The gap was ~7.5 percentage points per year — and the differential persisted after controlling for risk, market exposure, and other factors. The paper's title quotes the finding directly: trading IS hazardous to your wealth. The follow-up 2001 paper 'Boys Will Be Boys' showed that men, who traded ~45% more than women in the same dataset, also earned approximately 1.4 pp/yr lower net returns specifically because of the additional turnover. The 2008 day-trader study (Taiwan, 360,000 day traders 1992-2006) found that ~80% of day traders LOST money on a 12-month horizon and the most active 1% lost the most. Compounded over a 30-year investing career at 10% (passive) vs 4% (high turnover net of costs and bias-driven losses), the wealth gap is roughly 6x. The empirical case for trading less is one of the most robust findings in retail-investor research. Source: Barber & Odean, Journal of Finance, April 2000; follow-up 2001 and 2008 papers.
At every major market extreme, sentiment indicators hit extremes too. These are contrarian signals when paired with independent analysis. AAII Investor Sentiment Survey readings; values are the percentage of survey respondents bullish (peaks) or bearish (troughs).
The Superinvestors tab shows what the top fundamental managers did during major sentiment extremes — Buffett, Klarman, Marks, and others were typically buying during March 2009 and March 2020 when retail money was flowing out, and selling or hedging at speculative peaks like 2000 and 2021. The Institutional Flow tab shows the divergence between retail and institutional positioning in real time — useful as a contrarian signal when the two diverge sharply. The /macro page tracks the broad sentiment regime via market-internals and credit-spread data. Most importantly, the Stock Analysis page on every name shows you the company's current fundamentals (Overview, Ratios, Valuation, KPIs) — the only honest grounding for an investment decision. Whenever you find yourself wanting to buy a name because of social-media momentum, force yourself to compute fair value INDEPENDENTLY using the M6 framework before checking what the price is. If the M6 result and the current price are far apart, that's information; the disagreement should be the basis of the decision, not the social pain of missing out.
Two failure modes appear constantly in retail investing. First: chasing the rally. The investor who didn't own NVIDIA in early 2023 watched it triple by year-end, bought aggressively in early 2024 'because the AI thesis is real,' and then sat through 30%+ drawdowns later in the year. The thesis WAS real; the entry price was the problem. Buying after a major run with no margin of safety converts a structurally good investment into a structurally bad position. The discipline: compute fair value independently using M6's triangulation before committing capital, regardless of how good the story is or how many people are talking about it. Second: panic-selling the crash. The investor who had a quality 80/20 portfolio in February 2020, watched it drop 30% in 33 days, sold to cash 'until things stabilize,' and then waited too long to re-enter — missing the full recovery by August. Missing the 10 best days of the market over a 20-year window typically cuts long-run annualized return roughly in half (from ~10% to ~5% in commonly-cited Putnam Investments / Index Fund Advisors analyses). The 10 best days disproportionately occur immediately after the worst days — exactly when panic-sellers are out of the market. The discipline: do nothing during panics unless your written IPS (m7_l4) explicitly tells you to. Bias-driven action during extremes is statistically more likely to destroy returns than to preserve them.
We simply attempt to be fearful when others are greedy and to be greedy only when others are fearful. Most assuredly, fear and greed are in plentiful supply in the equity markets. Their occurrences also produce the prices that create the opportunities. We have never made a successful prediction about market direction. But we have, on a few occasions, recognized when other investors were generally either fearful or greedy — and acted in the opposite direction.