Retail Traders Lost 74-89% During Every Major Volatility Event: Study

As market volatility continues the largest-ever study of retail trading reveals the 27-year pattern behind their failure rate, and how AI can finally break the cycle of self-sabotage

DUBAI — AI trading lab PiP World has revealed the largest longitudinal study of retail trading in history, from 8 million traders, making 295 million trades across 27 years, pointing to one shocking conclusion. After nearly three decades the failure rate has not changed. A staggering 74% to 89% of retail investors still lose money, and always have. Retail traders have been losing the same way, for the same reasons, following the same four-phase spiral regardless of education, sophistication, trading app or cycle.

The stats haven’t moved and the reasons why traders lose haven’t either. Emotional self-sabotage has kept traders stuck in the same losing loop. PiP World’s big bet is that its new AI swarm can now intervene at the exact moment retail traders typically lose. Offering traders with a vital edge in all climates, in a week where sentiment turns bearish, or in the day-to-day volatility that now defines markets.

"Millions of trades almost read like poetry," said Saad Naja, Founder CEO of PiP World and veteran trader and former product leader at Kraken, Goldman Sachs, and Deutsche Bank. "It felt like rereading my own trading diary, the same rushes of greed, fear, and overconfidence, played out millions of times."

The four-phase spiral that predicts failure

The study identified something eerie: 85% of failed accounts followed the exact four-phase behavioral spiral consisting of: cautious success, overconfidence formation, catastrophic loss, and terminal decline. In the final phase, trading becomes compulsive, more emotional release than rational decision.

The behavioral analysis revealed when portfolios turn red, traders convince themselves they’re acting rationally even as discipline unravels. The four-phase spiral kicks in. Under pressure to recover, position sizes creep-up, stop-losses widen, and established risk rules are pushed aside. Trading exposes individuals to a potent mix of dopamine, ego, and stress. This combination results in emotional self-sabotage, overconfidence, revenge trading, fatigue, and poor timing repeat with clockwork precision.

“By the end, most traders are almost relieved when it’s over when they are out of their misery. The feeling of liquidation frees them from the constant stress and pressure of checking prices every 30 seconds,” Naja added.

Whilst the psycho analysis isn’t new, for the first time in history the solutions available to retail traders are. Retail traders no longer need to learn the triggers if they’re willing to let AI step in.

The repetitive four-phase spiral of 8 million traders over 295 million trades (Source: PiP World)

Enter the AI that trades discipline for dopamine

The failure rate has persisted over nearly three decades because “traders are humans who can’t ignore their emotions, plus trading app features are designed to feed on impulsivity,” Naja said, drawing on his candid experience as a former product leader. Humans don’t need to become emotionless to succeed. They only need to stop managing the part of the process where emotion does the most damage, the execution stage. Only when humans own the strategic mandate and cede execution, can they break the cycle of losses.

“Traders are safer steering strategy, leaving execution to the AI. It’s like choosing the destination in a self-driving car and enjoying the journey as a passenger. The roads and portfolios will be safer when the AIs aren’t navigating beside tired, distracted, road-raging emotional drivers,” Naja added.

PiP World is built on the strong conviction AI can help anyone to trade smarter. Unlike emotion-driven trading, which is almost always destined to lose. “Humans aren’t naturally disciplined,” said Naja. “So we built AI that is.”

Earlier this month, PiP World launched its answer to the decades-long cycle of loss and finally break the cycle of emotional self-sabotage: AI agents. A coordinated swarm of AI agents trained on what the best traders did and avoid what the worst traders did, from its enviable dataset of 275 million trades. A swarm designed to intercept failure at the execution stage where humans lose discipline, before it happens.

The AI agents work like a coordinated trading desk: a Chief Investment Officer agent orchestrates six specialist agents covering sentiment, fundamentals, technicals, macro, risk, and portfolio balance. The swarm uses continuous debate and evidence-based reasoning to converge on decisions that resemble institutional discipline.

The AI swarm aims to stem the losses, whilst you sleep

It’s the company’s first showcase of PiP’s proprietary AI Swarm Intelligence in the wild, marking the start of an ambitious roadmap toward AI-assisted live trading and agentic markets in 2026.

In markets that are inevitably stacked against retail because humans bring emotions and institutions are already bringing AI, PiP’s Swarm offers traders clarity, confidence, and control. It levels the playing field opening up access to elite fund-manager logic without needing to buy it, rent it, or understand it — like their portfolios depend on it.

All of PiP’s experiences are designed for easy entry, whilst building confidence and discipline in AI assisted trading at the users own pace. Requiring no prior knowledge of trading, crypto, or coding, PiP claims makes it far more accessible than the first movers.

"Our swarm doesn't just execute trades, it teaches anyone to think like institutional investors while handling the computational heavy lifting they could never do alone," Naja added.

As this trade shows, Naja’s been a driving force for PiP’s products to simulate trades before moving to live markets, but always deliver what paper trading has always lacked. Experiences that capture both the thrill and the discipline of high-stakes trading, so users build habits that translate directly into real-market performance.

The company recommends anyone exploring AI trading for the first time to begin by understanding the capabilities of AI agents, the trading behaviors they unlock, and benchmark performance – before deploying agents in live markets risking their money. That’s why PiP World has introduced ‘Market Mavericks’ in open beta, a sandbox simulation so anyone curious can test their AI agents, in real market conditions before risking any real capital. Each agent has its own model and trading behavior profile. They analyze real market data and simulate trades based on their trading personality.

By assigning three AI agents from a roster of 18 agents anyone can learn how AI agents trade crypto-pairs 24/7 for example about breakout detection (with Bit Wu) or macro-driven positioning (with Cath WoodFi), monitoring their P&L performance with full transparency. One early user described Market Mavericks as: “a sandbox that teaches fast and punishes slow thinking without bleeding capital,” @0xNhimZ35

AI agents simulating trades based on their unique trading personality from V1 Market Mavericks (Source: PiP World)

PiP World’s AI agents consistently delivered strong profitability under market chaos

The intense crypto sell-off last week provided a real-world stress scenario for PiP’s AI agents, and most demonstrated resilience by remaining profitable throughout. The agents continuously adapted to collapsing prices, vanishing liquidity, and extreme volatility in real time.

While the real markets were in freefall, PiP’s AI agents stayed calm, emotionless, and fearless. They showed diversified decision-making, each employing independent stop-loss and take-profit logic. Aside from the proof in their P&Ls, their trades showed remarkable composure. Each executed independent strategies that nonetheless aligned around profitable decisions. The essence of agentic intelligence, AI agents learnt to read disorder, adapt to uncertainty, and convert chaotic signals into coordinated, productive action.

P&L of AI agents based on virtual amounts users have invested, 21 Nov (Source: PiP World)

A similar pattern was observed during early testing prior to public launch, over the October 10 ‘crypto flash crash.’ The majority, 10 of 16 agents ended the week in profit, with a collective win rate of 66.2%.

"The flash crash gave us a glimpse into the future of markets," Naja said. "Each agent made independent decisions. Some cut risk, others waited for confirmation and a few leaned into the drawdown. Yet collectively they converged on profitable outcomes. What we’re seeing is the emergence of behavior that’s unique by agent. Evolving trading personalities shaped by real-time conditions, that's the essence of agentic intelligence."

PiP World’s AI performance during recent market turbulence proves emotionless, data-driven decision-making at speeds no trader can match delivers profits. All agent performance data is publicly verifiable and timestamped.

The dawn of intelligent wealth

The company's proprietary data moat, 27 years of institutional and retail trading data combined with real-time feeds powers continuously evolving AI agents that ingest over 7 million data points daily to identify structure in market noise.

While Version 1 is live to the public the models behind the AI agents are fast evolving, Version 3 is already in R&D the company teases offers groundbreaking updates with planned releases in the coming weeks. Aside from the AI infrastructure developments, more trading instruments, progression and rewards will roll out in future updates of Market Mavericks.

"The future of AI trading shouldn't be reserved for the first or the few. At the dawn of intelligent wealth powered by AI agents, everyone deserves a world-class hedge fund team in their pocket,” added Naja.

The company aims to transform retail investors from high-stakes gamblers into AI-assisted, VC-minded investors who allocate with intention, not impulse. Capturing the surging interest among retail investors turning to AI for investing, but harnessing it with informed conviction. Not blindly following the AI agent hype or underutilizing it as a context collector for surface level stock picks and sentiment analysis.

At the dawn of intelligent wealth, PiP’s Swarm Intelligence provides an exciting glimpse of a future where markets evolve from emotional battlegrounds into agentic ecosystems, where discipline, not dopamine, drives retail returns. One where AI swarms watch every market 24/7, spot risks, debate strategies, and execute without hesitation. Learning and acting faster than any human, spotting what humans miss, and never second-guessing themselves. In 2026, PiP World will give traders their first taste of AI-assisted live trading through expansion into its flagship product Co-Pilot.

About PiP World’s Survival Study

The study drew on 28 years of continuous trading data (1998–2025), covering 8 million trader profiles and 295 million from PiP World’s lead backer, Exinity retail trading platforms Alpari and FXTM. Tracking crypto CFDs, spot trading on equities, metals and ETFs through events including: 216 Federal Reserve meetings, 6 major market crashes (exceeding 20% drawdowns), and four full interest-rate cycles from zero-bound to 5%+ and back. Using survival-analysis models, analysts measured time-to-failure while isolating trader-driven decisions, leverage selection, trading frequency, loss-chasing, from external macro events. The patterns that increased failure rates emerged with clockwork precision.

Key Stats

• One of the largest known studies by dataset of retail trading data over decades: 8 million traders, making 295 million trades across 27 years
• 74–89% of retail traders lost money from 1998–2025: the persistent Failure Rate across all platforms, education levels, and regulations.
• Only 11–26% avoided losses over the same 27-year period: the Survival Rate.
• 85% of failed accounts followed the same Four-Phase Spiral behavioural pattern.

About PiP World

PiP World is an AI lab pioneering AI agents that give anyone the edge. The swarm of specialised AI agents compresses a high-performing trading desk into an intuitive AI trading experience that’ll guide any investors' next move. Transforming 295 million retail trades and real-time data into an agentic AI platform called ‘AI Swarm Intelligence’. PiP World’s ‘The Hedge Fund in Your Pocket’ AI experiences bridge the gap between institutional discipline and retail access. Built on 27 years of evidence. Designed to intercept failure before it happens. Turning market speculation into a structured, evidence-based collaboration between humans and AIs. At the dawn of intelligent wealth PiP World makes sophisticated AI trading accessible for all. Try Market Mavericks in open beta at pip.world/mavericks/play and follow @PiP_World, @MavAgents PiP.World, for the latest updates.