Copy Trading Psychology: Why Leaderboards Make You a Worse Trader

Risk disclaimer: Copy trading involves real financial risk. Past performance does not guarantee future results. Only trade with funds you can afford to lose entirely. This is not financial advice.

Copy Trading Psychology: Why Leaderboards Make You a Worse Trader#

Copy trading psychology biases are the systematic cognitive errors — survivorship bias, upward social comparison, and the disposition effect — that cause traders to select underperforming strategies based on leaderboard rankings rather than verifiable, risk-adjusted performance data. According to a 90-day study of 100,236 copy trading outcomes across Binance, Bybit, and MEXC (YieldFund, November 2025), only 48.48% of copy traders were profitable — even though 97% of the lead traders showed positive PnL on their own accounts. That gap is not bad luck. It is the predictable result of how leaderboards present information to human brains that evolved to follow successful-looking leaders.

This article explains — with academic research, regulatory findings, and on-chain data — exactly how leaderboard design exploits three cognitive biases, why the traders you copy are biased too, and what the structural alternative looks like.

You picked the number-one trader on the leaderboard. Their 30-day PnL was up 400%. Their win rate was 73%. Three weeks later, your portfolio is down 38%, and they have vanished from the rankings entirely. This is not an unusual story. According to one analysis of Binance copy trading data, 75% of copy trading portfolios close within four months. The leaderboard you used to choose that trader was not designed to help you make a good decision. It was designed to make you make a fast one.

The research is clear. The platform interface itself — the way it sorts, displays, and ranks traders — is an active participant in your losses.

The leaderboard is lying to you: survivorship bias in copy trading#

Survivorship bias in copy trading occurs when platforms display only active, currently profitable traders while failed or blown-up accounts silently disappear from the rankings. The result is that every trader visible on a leaderboard looks successful, because the unsuccessful ones are no longer there to see.

This is not a theoretical concern. According to Ammann, Burdorf, Liebi, and Stöckl (2022), who studied 3,904 cryptocurrencies over the 2014–2021 period, the annualized survivorship and delisting bias in crypto markets is 0.93% for value-weighted portfolios and 62.19% for equal-weighted portfolios. That second number means that if you only look at crypto assets that survived to the present day, their performance appears to be inflated by more than 62 percentage points per year compared to the full universe that includes failures.

Copy trading leaderboards replicate this exact distortion. A leaderboard sorted by "highest return" or "most copied" automatically filters out every trader who lost money and quit, every strategy that blew up, and every wallet that went inactive. What remains is a curated gallery of survivors — and their performance looks far better than it would if the failures were still visible.

Why is past performance not a guarantee in copy trading?#

Past performance on a leaderboard fails as a predictor because it reflects survivorship-filtered results, not a complete sample. The traders who lost money are no longer visible, making the survivors appear more skilled than probability alone would predict. According to the YieldFund study, 97% of lead traders recorded positive PnL on their own accounts, but only 43.61% of those leaders actually delivered positive returns to their followers.

This gap between leader profitability and follower profitability is the signature of survivorship bias in action — and a key reason why the question of whether copy trading is profitable depends entirely on how you select strategies. The leaderboard shows you the 97% who profited. It does not show you the universe of traders who tried, failed, and disappeared before you ever opened the app. By the time you see a trader's track record, it has already been pre-filtered by survival.

The same dynamic plays out in traditional finance. Mutual fund survivorship studies show that roughly 46% of funds are merged or liquidated over a 15-year period. The funds that remain — the ones you see in performance tables — look better precisely because the worst performers no longer exist.

Is a 100% win rate a red flag?#

Yes. A 100% win rate on a copy trading leaderboard almost always indicates one of three things: the trader has taken very few trades over a short period (luck, not skill), the trader is hiding unrealized losses in open positions that have not been closed yet, or the trader is running an account-boosting scheme.

Account boosting works like this: a trader opens four accounts and places opposite positions on each pair — two long, two short. After the outcome is known, the two losing accounts are abandoned. The two winners show 80% or higher returns and appear on leaderboards as top performers. This process can be repeated to create accounts with fabricated track records of 200% or more. The Daily Hodl documented this manipulation method in April 2025, noting that platforms rarely have mechanisms to detect it.

On centralized platforms where trading data is self-reported, this manipulation is difficult to detect. On-chain verification — where every trade is recorded permanently on a public blockchain and viewable through explorers like Solscan or Birdeye — makes fabrication structurally impossible. Every win, every loss, every position size is permanently recorded and independently auditable.

What leaderboards show	What leaderboards hide
30-day and 90-day PnL of active traders	All traders who lost money and stopped
Win rate (realized trades only)	Unrealized losses in open positions
Total follower count (social proof)	How many followers lost money
Headline ROI percentage	Drawdown depth and recovery time
Number of trades (activity signal)	Risk per trade and position sizing
Current rank position	How recently the trader reached that rank

Sources: leaderboard design analysis from IOSCO CR/10/2024 and Daily Hodl (April 2025)

Seeing top traders makes you trade worse: the social comparison trap#

The most direct evidence that leaderboards cause worse trading behavior comes from a 2023 experiment by Andraszewicz, Kaszás, Zeisberger, and Hölscher, published in Scientific Reports. The researchers recruited 807 experienced retail investors and had them trade on a simulated stock market. Half the participants were shown the results of top-performing traders. The other half were not.

The results were unambiguous. Traders who saw top performers' results took on significantly more portfolio risk (P < 0.001, effect size d = 0.2). Their trading volume increased by approximately 16.5% (P = 0.01). And their self-reported satisfaction with their own performance dropped substantially (P < 0.001, d = 0.36) — they felt worse about their results even when their actual returns were not meaningfully different.

This is upward social comparison at work. When you see someone else's exceptional performance, your brain does not process it as statistical noise. It processes it as information about what is achievable — and adjusts your behavior accordingly. You take bigger positions. You trade more frequently. You feel less satisfied with reasonable returns. All three responses make you a worse trader.

The effect is amplified in crypto markets, where social media accelerates the comparison cycle. According to the FINRA Social Media-Influenced Investing Report (December 2025), 72% of investors who rely on social media for investment decisions exhibited a far greater likelihood of taking on risky investments. Among investors under 35, 61% reported using finfluencer recommendations to make investment decisions. FINRA explicitly identified copy trading as a key strategy linked to social media platforms.

Is copy trading the same as gambling?#

Copy trading shares structural features with gambling — variable reinforcement schedules, 24/7 availability, social proof triggers, and the illusion of control — but it is not identical. The distinction matters because copy trading involves real assets with fundamental value, while gambling outcomes are purely probabilistic. However, a 2025 scoping review by Jain et al. across 13 studies and 11,177 participants found significant overlap between crypto trading behaviors and problem gambling patterns, including compulsive trading despite losses and social-media-driven herd behavior.

The gambling comparison becomes more apt when leaderboards are involved. A leaderboard that sorts by recent PnL and displays follower counts is functionally identical to a casino floor that shows only jackpot winners and hides the loss rate. The information environment is designed to trigger action, not analysis.

The International Organization of Securities Commissions (IOSCO) warned in its 2024 report on copy trading practices that these platforms are "usually promoted as simple and profitable despite the potentially complex and risky nature" and noted "poor investor outcomes and excessive risk taking due to ease of access for inexperienced investors."

The traders you copy are biased too: the disposition effect problem#

The biases do not stop with the follower. The traders being copied are also psychologically compromised by the leaderboard system — and their distorted behavior flows directly into your portfolio.

Pelster and Hofmann (2018) studied proprietary eToro trading data and found that becoming a signal provider — gaining followers who copy your trades — significantly increases the disposition effect. The disposition effect is the well-documented tendency to sell winning positions too early and hold losing positions too long. Signal providers exhibit this bias more strongly than non-followed traders, and the effect intensifies as the number of copying investors grows.

The mechanism is reputational fear. When followers are watching, realizing a loss feels like a public admission of failure. The signal provider holds the losing position longer, hoping it recovers, because closing it would reduce their displayed PnL and potentially cost them followers and fee income. The followers, meanwhile, inherit these underwater positions without understanding that the hold decision was driven by reputation management rather than market analysis.

What happens if the trader I copy loses money?#

When a signal provider loses money, their followers absorb the same percentage loss — but with an additional psychological cost. The follower did not make the trade decision and has no thesis to evaluate, so they cannot rationally assess whether the loss is within strategy parameters or signals a fundamental breakdown. This often leads to panic-exiting at the worst time or, conversely, staying too long out of misplaced trust in the leader.

Research published in Management Science by Apesteguia, Oechssler, and Weidenholzer (2020) provides the sharpest finding on this topic. In the first laboratory study of copy trading, they found that copy trading reduces ex-ante welfare and leads to excessive risk-taking. Previously successful traders — the ones who appear at the top of leaderboards — may simply have been lucky. Copiers then imitate these lucky traders, adopting strategies that are riskier than what they would have chosen independently.

The paradox is striking: risk-averse traders are the most likely to become copiers (because copy trading feels safer than independent trading), but copying leads them to adopt riskier strategies than they would have chosen on their own — which is why proper risk management matters more in copy trading than in independent trading.

A separate 2025 study by Pelster in the Journal of Behavioral and Experimental Finance confirmed that traders with increased social audience size trade more frequently, use higher leverage, and attain poorer performance — and that these adverse effects intensify among traders who previously excelled. The very act of being watched makes good traders worse.

How leaderboard UI design amplifies every bias#

The biases described above are not accidents. They are the predictable outcomes of specific user interface design choices that almost every copy trading platform makes. Understanding which design element triggers which bias is the first step toward making better decisions — or choosing platforms that do not rely on these mechanisms.

Design element	Cognitive bias it triggers	How it harms you
Sort by highest recent PnL	Survivorship bias + recency bias	Shows only current winners; hides blown-up accounts and long-term underperformers
Follower count badge	Herding / social proof	Makes popular traders appear safe; popularity is not correlated with future performance
Win rate percentage (realized only)	Anchoring + selective disclosure	Excludes unrealized losses; a 90% win rate with massive hidden drawdowns is worse than 50% with controlled risk
"Top trader" or "Elite" label	Authority bias	Platform-assigned status signals expertise without verification of methodology or risk management
30-day performance window	Recency bias	Overweights recent luck; statistically meaningless for evaluating long-term edge
No drawdown data shown	Omission bias	What is not shown is assumed to be fine; platforms have no incentive to display risk metrics prominently

Adapted from IOSCO's analysis in CR/10/2024 and behavioral finance frameworks from Barber and Odean (2000)

Platforms have limited incentive to change these design patterns. Leaderboards drive engagement. Engagement drives trading volume. Trading volume drives fee revenue. A platform that charges 1% per trade — the industry standard on Solana across platforms like GMGN, Axiom, Trojan, and BullX — earns more revenue when users trade more frequently. The 16.5% increase in trading volume caused by social comparison (Andraszewicz et al., 2023) is not a bug. It is a revenue feature.

What metrics should I look at when choosing a trader to copy?#

Instead of relying on the two metrics leaderboards prioritize (recent PnL and follower count), evaluate traders across at least six dimensions: win rate across a minimum 100-trade sample, maximum drawdown from peak equity, profit factor (gross profit divided by gross loss), consistency of returns across different market conditions, average position size relative to portfolio, and the ratio of realized to unrealized PnL. If a platform does not display all six, it is hiding information that affects your decision.

The strongest possible verification is on-chain. Platforms built on public blockchains like Solana allow every trade to be independently audited through block explorers like Solscan. If a platform cannot provide a verifiable wallet address for each strategy, the performance data is self-reported — and self-reported data in a system with financial incentives to look profitable is, by definition, unreliable.

What algorithmic curation looks like — and why it works#

The alternative to leaderboard-based selection is algorithmic curation: replacing human judgment and UI-driven social signals with multi-dimensional scoring systems that evaluate strategies on data humans tend to ignore.

Instead of ranking traders by a single metric (recent PnL) and letting social proof (follower count) determine visibility, an algorithmic approach — like the one described in our guide on how copy trading works on Solana — scores wallets across multiple dimensions simultaneously: edge quality, consistency across market conditions, drawdown stability, risk-adjusted return sharpness, asset diversity, whether the strategy's trade sizes are replicable at follower scale, and position sizing discipline.

This approach directly addresses each bias the leaderboard amplifies. Survivorship bias is eliminated because the scoring system analyzes full trading history — including periods of loss — rather than filtering by current profitability. Social comparison is removed because there is no ranking, no follower count, and no "top trader" badge. The disposition effect is detected because drawdown metrics and holding-period analysis reveal when a trader is sitting on underwater positions instead of cutting losses.

On-chain verification provides the trust layer. When every trade — including losses — is recorded permanently on a public blockchain like Solana and viewable through Solscan, performance data cannot be fabricated. Account boosting schemes fail because the opposite-position accounts are visible. Unrealized PnL hiding fails because open positions are on-chain.

Stratium is a Solana copy trading platform built on this model. It charges 0.1% per trade — compared to the 1% industry standard — and scores strategies algorithmically across multiple risk-adjusted dimensions rather than ranking them on a leaderboard. Every trade, including losses, is publicly verifiable on Solscan. Three of the platform's eleven current strategies show negative PnL, and those losses are displayed on the homepage alongside the winners. That is the opposite of survivorship bias.

Verify the claim yourself. Every strategy wallet's complete trade history is on Solscan — you can audit every entry, every exit, every loss before you deposit a single SOL.

How long should a trader's track record be before I copy them?#

A trader's track record should span at least 100 trades across a minimum of three months — and ideally through at least one significant market drawdown. Short track records in favorable market conditions tell you almost nothing about skill versus luck. According to Apesteguia et al. (2020), previously successful traders on copy trading platforms may simply have been fortunate, and copiers who imitate them are adopting strategies that survived by chance rather than by edge.

The strongest evidence of genuine skill is consistent performance through varying market conditions — up, down, and sideways. If a strategy has only been tested during a bull run, you are copying luck, not methodology. Platforms that evaluate strategies algorithmically across multiple market regimes filter for this automatically. Leaderboards, by design, do not.

FAQ#

Why do most copy traders fail?#

Most copy traders fail because they select strategies based on leaderboard rankings that are distorted by survivorship bias, social comparison triggers, and selective performance display. According to a YieldFund study of 100,236 outcomes, only 48.48% of copy traders were profitable — even though 97% of the lead traders showed positive PnL on their own accounts. The gap exists because leaderboards show you the survivors, not the full sample.

What percentage of copy traders make money?#

Across Binance, Bybit, and MEXC, approximately 48.48% of copy trading outcomes were profitable over a 90-day period, according to YieldFund (November 2025). Results varied significantly by platform: Binance at 66.5%, MEXC at 57.79%, and Bybit at 43.65%. For context, eToro's own regulatory disclosures state that 67% of retail investor CFD accounts lose money.

Can you verify a copy trader's performance on-chain?#

Yes — if the platform operates on a public blockchain like Solana. Every trade executed on Solana is permanently recorded and viewable through block explorers such as Solscan or Birdeye. This means you can independently audit a strategy's complete trade history, including losses, position sizes, and timing, without relying on the platform's self-reported data. Platforms that cannot provide a verifiable wallet address for their strategies are asking you to trust screenshots instead of blockchain records.

What is the disposition effect and how does it affect copy trading?#

The disposition effect is the tendency to sell winning positions too quickly and hold losing positions too long. In copy trading, Pelster and Hofmann (2018) found this bias intensifies for signal providers once they gain followers, because realizing a loss feels like a public failure. Followers inherit the consequences: they end up holding underwater positions that the signal provider is too reputation-conscious to close.

The leaderboard is the product, and you are the customer#

The central finding across all the research cited in this article is consistent: leaderboard-based copy trading platforms create environments where cognitive biases are not just present but actively amplified by interface design. Survivorship bias filters out failures. Social comparison increases risk-taking and reduces satisfaction. The disposition effect intensifies under observation. And the UI design choices that cause these outcomes also happen to increase trading volume — which increases platform revenue.

The solution is not better willpower, more emotional discipline, or a deeper understanding of your own psychology. Those are individual fixes for a structural problem. The solution is infrastructure that removes the bias triggers entirely: algorithmic scoring instead of leaderboards, on-chain verification instead of self-reported statistics, and full loss transparency instead of selective display.

Every Stratium trade — including losses — is verifiable on Solscan. The platform does not rank strategies by PnL or display follower counts. It scores them algorithmically. You do not need to trust the ranking. You can verify every trade yourself.

---

This article is for informational purposes only and does not constitute financial advice. Crypto trading involves substantial risk of loss. Past performance does not guarantee future results. The academic studies cited above were conducted in controlled or observational settings and may not directly predict individual outcomes in live crypto markets.

---

About the author: Florian is the founder of Stratium, a copy trading platform on Solana where every trade is publicly verifiable on-chain. He trades his own capital in the same strategies available to users. Strategy wallet history is on Solscan.