| Research thesis Copy trading can outsource decision-making, but it cannot outsource market structure. The more capital converges on the same visible leaders, instruments and styles, the less independent those strategies become. A copied portfolio may show five or ten different names while still carrying one underlying bet: the same momentum factor, the same technology stocks, the same crypto beta, the same dollar trade, or the same leveraged exit. Crowding can therefore turn apparent diversification into synchronized order flow – and make past success less portable into the future. |
Key findings
| Finding | Why it matters |
| Copying creates an order-flow multiplier | A leader trade can become a much larger synchronized order when copied across millions of dollars of follower capital. eToro itself warns that concentrated copying can create settlement and market-integrity risks. |
| Number of leaders is not the same as diversification | Five copied traders can still be one portfolio if they share the same exposures. Correlation matters more than the number of profile names. |
| Popularity can arrive after performance | A 2026 study of 500 certified traders and 167 popularity jumps found strong abnormal returns before the popularity event, but no significant persistence afterward. |
| Leverage makes crowding nonlinear | When correlated strategies use leverage, the same price shock can hit margin thresholds and stop-outs at roughly the same time, turning common exposure into common forced selling. |
| Leaderboards contain survivorship bias | Visible leaders are selected from past outcomes. Failed, dormant or de-ranked strategies receive less attention, making the displayed opportunity set look cleaner than the full strategy population. |
| Risk controls help, but do not remove common-factor risk | Copy stop-losses, allocation caps, risk scores and suitability checks can limit damage. They cannot guarantee execution or prevent several copied traders from making the same economic bet. |
The real product is not the trader – it is synchronized flow
Copy trading is often described as delegation: a retail investor finds someone with a credible track record and automatically replicates the trades. That description is mechanically accurate, but it misses what happens after a leader becomes popular. The leader is no longer only managing one portfolio. The leader is effectively emitting a trading signal that can trigger orders across many other accounts at almost the same moment.
eToro gives a rare public indication of the scale this can reach. Its 2025 annual filing says the platform had more than 4,750 members of its Pro Investor program at year-end. Seventeen had more than $10 million each in Assets Under Copy, while 125 had more than $1 million. The same filing explicitly warns that if a large number of investors copy a single Pro Investor, the volume of simultaneous trades can create operational problems and even failed settlement relative to the availability of the asset in the market.
That disclosure is important because it converts “crowding” from a behavioral concept into an execution problem. A leader might decide to reduce a position by 5%. If the leader has $250,000 of personal capital, the leader-only sale would be $12,500. If another $10 million is proportionally copying the strategy, the synchronized economic flow implied by the same 5% rebalance becomes roughly $512,500 – about 41 times the leader-only order in this illustration. At $25 million of copied capital, the same signal implies more than $1.26 million of synchronized notional.
This does not mean every platform routes orders identically or that all copied trades reach the market at the same price and instant. It does mean that the scalability of a strategy depends on more than whether the leader can trade it successfully in a small account. Capacity becomes part of the investment thesis.
Figure 1. Illustrative copy-flow amplification. eToro disclosed that 17 Pro Investors had more than $10 million each in Assets Under Copy at year-end 2025. The leader account size and rebalance percentage shown here are illustrative, not eToro trading data.
Diversification can disappear without the investor noticing
The most common copy-trading mistake is to diversify by personality rather than by exposure. An investor may copy five leaders who have different biographies, trading styles and profile statistics, yet all five can be long the same underlying risk factor. One may own Nvidia directly, another may trade Nasdaq futures, a third may hold a technology ETF, a fourth may own Bitcoin as a high-beta liquidity trade, and a fifth may be long a growth basket. The positions look different; the macro bet can be almost identical.
The mathematics is unforgiving. If five equally weighted strategies each have 10% standalone volatility and are genuinely uncorrelated, the combined volatility is about 4.5%. If their average correlation rises to 0.50, the same portfolio volatility rises to about 7.7%. At perfect correlation it returns to 10% – effectively no volatility diversification at all. The important variable is therefore not how many people are copied, but how independent their return streams really are.
This is why a platform-level risk score and a trader-level maximum drawdown are not enough. Correlation is a portfolio property. A trader with a moderate individual risk score can still increase the risk of the copier if that trader owns the same exposures as the rest of the copier portfolio. eToro has recognized this problem in product design: its portfolio-insight tools show which assets contribute most to portfolio risk and identify assets with lower correlation to current holdings.
Figure 2. Correlation can erase the diversification benefit of copying multiple traders. The chart is a mathematical illustration, not historical platform performance.
Leverage turns common exposure into common liquidation risk
Correlation becomes more dangerous when copied strategies use leverage. The first effect is obvious: the same market move produces a larger percentage change in equity. The second effect is more important for crowding: leveraged portfolios have thresholds. Margin calls, stop-outs, copy stop-losses and trader-defined exits can all convert a gradual market move into forced trading.
If several copied leaders are long the same theme with leverage, a shock can push many accounts toward their risk limits simultaneously. The first wave of selling can worsen execution for the next wave, particularly in thinner instruments or outside the most liquid trading hours. This is the copy-trading version of a familiar market-structure problem: leverage does not merely magnify losses; it can synchronize the moment when positions must be reduced.
eToro says it generally recommends that users place no more than 20% of equity in one trade and allocate no more than 40% of total balance to copying one trader. Its Copy Stop Loss mechanism can also close an entire copy relationship after account equity reaches a user-set threshold. Those are useful controls, but the company also warns that stop orders are not guaranteed to execute at the specified level. In a fast market, a risk control is a rule for attempting an exit, not a guaranteed exit price.
Strategy decay: success attracts the capital that can weaken success
The paradox of copy trading is that visibility is awarded for performance, but visibility can change the conditions under which that performance was produced. A low-turnover strategy in liquid large-cap stocks may absorb additional follower capital with little effect. A fast strategy trading smaller stocks, crypto pairs or short-lived price dislocations may not. Once copied capital becomes large relative to the strategy capacity, slippage, delayed fills and adverse selection can eat into the edge.
A 2026 Financial Research Letters study provides unusually direct evidence on the popularity problem. Using proprietary eToro data covering 500 certified traders and 167 sudden popularity jumps, the researchers found significant abnormal returns before the popularity event, but post-event returns turned slightly negative and were statistically insignificant. Their matched-control analysis likewise found no return advantage from selecting a trader simply because that trader had become trendy.
The implication is not that popular traders must underperform. It is that popularity is an endogenous signal: traders usually become popular because something has already gone right. Capital therefore arrives after the event that generated the visibility. The copier is buying the persistence of a historical pattern, not the historical pattern itself.
A separate 2025 study of social-trading records found that stronger peer performance increased investors’ trading activity; the subsequent increase in activity was associated with lower performance and higher volatility. That is a behavioral version of strategy decay: success in the network can stimulate more risk-taking precisely when the social signal is strongest.
Survivorship bias makes the leaderboard look safer than the strategy population
Every copy-trading leaderboard is a filtered sample. Traders who blew up, stopped trading, breached platform rules, lost followers or simply became inactive are less visible than the current cohort of leaders. The investor therefore observes the winners who survived long enough to qualify for attention, not the full distribution of people who attempted the same strategy.
This creates two distinct biases. The first is conventional survivorship bias: failed strategies disappear from the comparison set. The second is selection-on-recent-performance: the very act of becoming visible often depends on strong historical returns, controlled drawdowns or high follower interest. If returns contain luck as well as skill, ranking by realized performance mechanically pushes unusually lucky recent histories toward the top.
eToro addresses this directly in its risk disclosures. It says past performance, risk scores and other statistics should not be treated as guarantees of future results, and it cautions that a copied trader’s future risk may exceed the displayed historical risk score. ESMA goes further in its supervisory guidance: past performance should be fairly represented, should not be the most prominent element of marketing, and firms should explain how performance figures are calculated.
The retail lesson is simple: a leaderboard is a discovery tool, not a due-diligence process. The right question is not “Who has the highest return?” but “What economic exposures produced the return, how long has the process worked, how much capital now follows it, and what would cause it to stop working?”
Figure 3. The leaderboard selection trap: investors usually allocate after a trader has already survived a performance and visibility filter.
There is also an incentive problem inside social trading
Popular leaders can be paid for being copied. eToro says members of its Pro Investor program can receive compensation that rises with copy engagement and, at higher tiers, can be linked to a percentage of Assets Under Copy. This aligns the leader with building a durable following, but it can also create a potential tension between gathering assets and preserving strategy capacity.
European regulators have noticed the issue. ESMA’s copy-trading briefing specifically asks firms to examine whether copied-trader remuneration depends on the number or value of copied trades, follower counts or average assets under management, and whether those incentives could cause the copied trader to favor their own interests over the clients’ interests. The regulator also says firms should manage conflicts and ensure suitability or appropriateness depending on how the copy service is structured.
This does not make compensated leaders inherently conflicted. Traditional asset managers also earn more when assets grow. The difference is that copy-trading platforms can make asset gathering visible and socially reinforced in real time: follower counts, performance rankings and public discussion all feed the same attention loop.
When copy trading is most likely to survive its own popularity
Copy trading is not structurally doomed. Some strategies are naturally scalable. Long-horizon portfolios in deep, liquid instruments can absorb substantial copied capital with little market impact. Low turnover reduces the importance of millisecond execution differences. Moderate leverage lowers the probability of synchronized forced exits. Transparent factor exposure makes it easier for the copier to identify overlap across leaders.
The most robust setup is therefore almost the opposite of what social-media incentives tend to reward. The safer candidate is usually not the trader with the most spectacular recent return, but the trader with a long history, explainable exposures, controlled leverage, low dependence on thin instruments, stable turnover and enough strategy capacity relative to the capital already copying the account.
Risk controls should also exist at the copier portfolio level. A user can cap the percentage allocated to any one leader, but should also cap exposure to a single theme across all leaders. A 20% allocation to five different traders is not diversified if all five are effectively long the same risk factor. Correlation, gross leverage, overlapping instruments, turnover and assets-under-copy capacity are more informative than follower count alone.
A practical risk-control stack for copy traders
| Risk | Control to look for | What it cannot solve |
| Crowding / capacity | Assets Under Copy limits; copy caps; avoid thin instruments and very high-turnover leaders | Does not stop many leaders from owning the same asset or factor |
| Correlation | Portfolio-level exposure breakdown; pairwise-return correlation; sector/factor limits | Historical correlation can jump toward 1 during stress |
| Leverage | Gross leverage caps; position-size limits; margin buffers | Cannot guarantee exits during gaps or fast markets |
| Drawdown | Copy stop-loss; maximum allocation per leader; portfolio loss budget | Stop price may differ from actual execution price |
| Strategy decay | Longer history; capacity analysis; turnover monitoring; compare pre- and post-popularity behavior | A once-valid edge can still disappear |
| Survivorship bias | Review full history, inactive periods, max drawdown, tenure and benchmark-relative returns | Platform screens may still omit failed or departed peers |
| Incentive conflict | Understand how leaders are paid and whether compensation rises with followers/AUC | Disclosure does not remove the incentive itself |
| Retail takeaway Copy trading should be treated as portfolio construction, not personality selection. Before adding a new leader, ask whether the trade stream is genuinely independent from the leaders already copied. A lower-return strategy with different exposures can reduce portfolio risk more effectively than another high-return strategy that behaves exactly like the existing book. |
Eight questions to ask before copying a trader
1. What percentage of the leader’s return came from one asset, sector, currency or crypto theme?
2. How correlated is this leader with the other people I already copy?
3. How much leverage is used, and can leverage change without my approval?
4. How much capital is already copying the strategy relative to the liquidity of what it trades?
5. Is performance driven by frequent trading where slippage matters, or by slower portfolio allocation?
6. How did the strategy behave during its worst month, not just its best year?
7. How long is the verified history, and did the leader become popular only after a short winning streak?
8. How is the copied trader compensated – by performance quality, follower count, Assets Under Copy, trading activity, or some combination?
The bottom line
Copy trading solves a genuine retail problem: many investors do not have the time, confidence or expertise to build and manage every position themselves. A well-designed copy system can make experienced decision-making more accessible and can expose users to strategies they would not otherwise know how to implement.
But the system changes as it succeeds. The most visible traders attract the most assets; the most copied trades create the most synchronized flow; and the same recent performance that draws followers can be the performance least likely to repeat. Once leverage and overlapping exposures are added, a portfolio of many copied traders can become one crowded trade wearing several usernames.
The durable edge, if there is one, is therefore not copying the highest-ranked trader faster than everyone else. It is understanding the hidden portfolio created by all the people being copied – and managing correlation, capacity, leverage and incentives before the crowd discovers the same risk at the same time.
Methodology and limitations
This article combines primary regulatory and corporate disclosures with peer-reviewed academic research on social trading and retail herding. Platform-specific examples are used to explain mechanisms, not to claim that every copy-trading provider uses identical execution, ranking or compensation systems.
The copy-flow amplification chart is an illustrative calculation based on proportional copying. The $250,000 leader balance, 5% rebalance and example AUC levels are assumptions; only the fact that some eToro Pro Investors had more than $10 million in Assets Under Copy is taken from the company’s filing. The correlation chart is a standard equal-weight portfolio calculation using assumed 10% standalone volatility. Neither chart is a forecast or measured platform return series.
Academic results cited here describe specific datasets and should not be generalized mechanically to every market, platform or trader. Copy trading can involve stocks, ETFs, cryptoassets, CFDs, futures and other instruments, each with different liquidity, leverage and regulatory characteristics.
Sources
1. 2025 Form 20-F – eToro / U.S. SEC. Operational and regulatory risks, Pro Investor counts, Assets Under Copy thresholds, compensation structure and platform disclosures. Source link
2. CopyTrader Guide, January 2026 – eToro. Mechanics of proportional copying, risk-number context and account execution explanation. Source link
3. CopyTrading Risks – eToro. Past-performance, risk-score and stop-order limitations. Source link
4. Responsible Trading Policy – eToro. Suggested leverage, position-size and copy-allocation limits. Source link
5. Copy Systems Explained – eToro. Copy Stop Loss mechanics and default threshold. Source link
6. Portfolio tools and diversification features – eToro. Portfolio breakdown, risk insights and correlation-oriented diversification tools. Source link
7. Supervisory Briefing on Copy Trading – ESMA. MiFID II expectations on marketing, suitability, appropriateness, remuneration, conflicts and copied-trader oversight. Source link
8. When popularity strikes: Evidence from an event study on social trading platforms – Financial Research Letters, 2026. Study of 500 certified traders and 167 popularity jumps; no significant post-popularity return persistence. Source link
9. The impact of peer returns in social trading – Journal of Behavioral and Experimental Finance, 2025. Peer performance, increased trading activity, subsequent performance and volatility. Source link
10. The correlated trading and investment performance of individual investors – Journal of Empirical Finance, 2024. Evidence linking retail herding with weaker performance and execution disadvantages. Source link
11. The role of herding in the rise of leader traders on social trading platforms – Computers in Human Behavior Reports, 2026. Experimental evidence that social signals can induce herding in copy-trader selection. Source link
12. ESMA Q&A 2463 – copy trading under MiCA – ESMA. Application of copy-trading service qualification concepts to crypto-asset services under MiCA. Source link
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape.