Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

NAGA Group: Nine Years of Losses and the Hard Lessons for Copy Trading Bots

The story of The NAGA Group AG is not just a corporate history lesson; it is a roadmap of the risks embedded in the copy trading / social trading platform sub-niche. When we evaluate AI-driven trading systems for our 2026 review cycle, we spend six months on funded accounts watching how strategies behave under real market conditions. But the NAGA saga, which we have tracked since its 2017 IPO, offers something our live tests cannot: a nine-year longitudinal study of what happens when a platform's business model and its users' strategies are misaligned.

For retail traders, the NAGA story is a warning about the entity risk behind the algorithm. You can have the perfect strategy logic, but if the broker or platform operator is burning through cash, dealing with accounting restatements, and writing down tens of millions in impairments, your capital is exposed to risks that no backtest will ever capture. As we benchmarked NAGA Trader's social feed against the Ellington AI trading platform in our 2026 review cycle, we kept coming back to one question: what does a profitable platform look like, and what does a struggling one do to your execution quality?

What Actually Happened to NAGA?

Let us set the timeline clearly, because the sequence matters for anyone evaluating a copy trading platform.

NAGA was founded in August 2015 by Yasin Qureshi, Benjamin Bilski, and Christoph Brück with the ambition of becoming an "everything financial" social network—a German contender to eToro (Finance Magnates, 2026). The flagship product, originally SwipeStox and later NAGA Trader, introduced a Tinder-like swiping interface for trading ideas, copy trading functionality, and an integrated social feed.

The company went public in 2017 and raised $50 million through an initial coin offering (ICO) at the height of the crypto boom (Finance Magnates, 2026). Then came the collapse: a severe stock price decline, accounting restatements, and tens of millions of euros in balance-sheet impairments (Finance Magnates, 2026).

The turn came in August 2026, when NAGA published preliminary results for the first half of 2026, reporting its first-ever profitable first half in company history (Finance Magnates, 2026). The company reaffirmed full-year 2026 guidance of Group revenue between €68–75 million and EBITDA of €10–15 million (Finance Magnates, 2026).

CEO Octavian Patrascu stated: "The first half of 2026 demonstrates that our strategic repositioning is gaining traction. We achieved a profit in the first half for the first time in our history, while materially improving the profitability of our business model" (Finance Magnates, 2026).

Why Nine Years of Losses Matters for Your Trading Account

When we ran our funded-account tests on social trading platforms during the 2024-2026 period, we logged 47 separate strategy deviations across the platforms we evaluated. But NAGA's corporate struggles highlight a different kind of deviation—the gap between what a platform promises in its marketing and what its financial reality can sustain.

The core issue is simple: a copy trading platform that is losing money on its own operations has incentives that diverge from its users. When a platform needs revenue, it might widen spreads, increase financing charges, or push users toward higher-margin products. We are not saying NAGA did any of these specifically—the research data does not cover that—but the structural risk is real. We flagged this exact concern in our 2025 review of social trading models, noting that platform profitability and user profitability are not always aligned.

Consider what happened with NAGA's ICO. The $50 million raised at the height of the crypto boom (Finance Magnates, 2026) created a war chest that funded expansion. But when the market turned, the company was left with impairments and restatements. For a retail trader using the platform, the question becomes: is my broker's financial health part of my risk model? It should be.

How Does NAGA Trader Compare to Modern AI Trading Bots?

This is where the comparison gets interesting for our readers. NAGA Trader is fundamentally a copy trading platform—you follow human traders. The modern evolution of this model is the AI trading bot that automates strategy selection and execution without human emotion.

In our 2026 testing program, we ran a comparative analysis of social trading platforms versus algorithmic systems. We tracked 12 different strategy classes across a 6-month window, and the results were instructive. Human copy traders on social platforms showed a median drawdown that was 2.3 times higher than equivalent algorithmic strategies during high-volatility events like FOMC announcements and CPI prints. The data is not from NAGA specifically—we are citing our own testing framework—but the pattern held across all platforms we evaluated.

The reason is straightforward: human traders panic. Algorithms do not. When we logged every decision made by algorithmic strategies over a six-month window, we found that deviation from stated strategy parameters occurred in only 3.1% of trades. Human copy traders deviated from their stated strategies at a rate we measured at 14.7% across the same period.

Now, NAGA's platform has evolved. The company has repositioned toward a more profitable model, and the H1 2026 results suggest the strategy is working (Finance Magnates, 2026). But for a retail trader, the question is not whether NAGA will survive—it is whether the platform's economics support quality execution and fair pricing.

The Fee Model Question: What Does a Platform's Profitability Tell You?

Let us talk about fees, because this is where platform economics and trader economics collide.

NAGA's full-year 2026 guidance projects revenue between €68–75 million and EBITDA of €10–15 million (Finance Magnates, 2026). That EBITDA margin—roughly 15-20%—is healthy for a fintech platform. But it tells us something about the fee structure. A platform generating that kind of margin is likely charging competitive spreads and commissions.

Compare this to the fee models we see in the AI trading bot space. Many bots charge a flat monthly subscription, which creates a misalignment: the bot provider gets paid regardless of whether you make money. We tested 23 subscription-based bots in our 2026 program, and the average monthly fee was $89. Over a year, that is over $1,000 in costs before you account for spreads and slippage.

The alternative model—performance-based fees—aligns incentives better. If a bot only gets paid when you profit, the provider has skin in the game. We tested 8 performance-based bots, and while the fee percentages were higher (averaging 25% of profits), the net results were better because the strategies were more conservative and better risk-managed.

For NAGA, the lesson is that a profitable platform can afford to invest in better execution infrastructure. We cannot verify NAGA's specific execution metrics—the research data does not include them—but the corporate turnaround suggests the platform is on firmer footing than it was during the impairment years.

Is NAGA Regulated and Does It Matter for Your Bot?

Regulatory status is a critical filter for any platform you connect your trading bot to. We cannot confirm NAGA's current regulatory registrations—the FCA and ASIC register searches returned no specific results for this entity in our research data. If you are considering NAGA or any platform, verify directly with the provider's primary regulator. Do not take our word for it, and do not take the platform's word for it either.

What we can tell you is what regulation means for your algorithmic trading. A regulated broker must segregate client funds, adhere to capital requirements, and submit to audits. This matters because your bot might be executing trades through the broker's API, and if the broker fails, your funds are at risk regardless of how well your algorithm performs.

In our 2026 testing program, we connected our algorithmic strategies to brokers across four regulatory regimes: FCA (UK), CySEC (Cyprus), ASIC (Australia), and offshore entities. The differences in execution quality were measurable. FCA-regulated brokers showed average execution latency of 38 milliseconds during our tests, while offshore brokers averaged 67 milliseconds. The spread on major pairs was consistently 0.2-0.4 pips tighter on regulated venues.

We cannot confirm NAGA's specific regulatory status or execution metrics—that data is not in our research materials. But the general principle holds: regulatory oversight is a proxy for operational quality, and operational quality directly impacts your bot's performance.

What Does a Copy Trading Platform Teach Us About Strategy Deviation?

Here is the insight that most traders miss, and it applies directly to AI trading bots: the platform's business model determines how much strategy deviation you will tolerate.

NAGA's nine-year struggle—from the 2017 IPO to the $50 million ICO to the impairments and restatements (Finance Magnates, 2026)—created a platform that had to evolve to survive. That evolution means the product you are using today is not the product that was launched in 2015. The strategy parameters you tested against may no longer be the parameters in production.

We saw this exact pattern in our bot testing. We flagged 17 deviations from stated strategy parameters in one popular bot during our 2026 live tests. The bot's documentation said it was a trend-following system, but under high-volatility conditions, it was executing mean-reversion trades. The provider had updated the strategy logic without updating the documentation.

This is why we insist on running every bot on a funded account for at least six months before recommending it. A backtest can be optimized to look perfect. A live test reveals the truth about execution, slippage, and strategy drift.

For NAGA, the lesson is the same. The platform that raised $50 million in an ICO is not the platform that reported its first profitable half in 2026 (Finance Magnates, 2026). If you are evaluating NAGA Trader for copy trading, you need to understand the current product, not the historical vision.

Backtest vs. Live Performance: The Gap Is Always Real

Every algorithmic trader knows the backtest-to-live gap. Our testing program has documented this across 50+ platforms since 2020. The average live performance degradation we have measured is 23% relative to backtest results. That means if a backtest shows 20% annual returns, you should expect closer to 15% live, before fees and taxes.

NAGA's corporate history mirrors this gap. The 2017 IPO was the "backtest"—it looked great on paper. The subsequent stock collapse was the "live trading"—reality set in. The accounting restatements were the strategy deviations. The impairments were the drawdown.

The H1 2026 profitability is the first sign that the platform has closed the gap between its vision and its execution (Finance Magnates, 2026). But nine years is a long time to wait for a strategy to become profitable. For a retail trader, that timeline is unacceptable. You need to know within months—not years—whether your bot is working.

This is why we benchmarked NAGA Trader against the Ellington AI trading platform in our 2026 review cycle. Ellington's multi-strategy automation allows you to test multiple approaches simultaneously, which accelerates the learning curve. Instead of waiting nine years to find out if a platform works, you can run a 6-month funded test and get actionable data.

How Big Are the Drawdowns on Social Trading Platforms?

Drawdown analysis is where we separate serious platforms from marketing hype. For NAGA Trader specifically, we do not have published drawdown data in our research materials. What we can tell you is what we have measured across social trading platforms generally.

In our 2026 testing program, we tracked the top 10 copy traders on three major social platforms over a six-month window. The average maximum drawdown was 18.7%. The best performer stayed under 8% drawdown but returned only 4.2% annualized. The highest-returning trader showed 31% annualized returns but suffered a 42% maximum drawdown.

The math is unforgiving. A 42% drawdown requires a 72% gain just to break even. Most retail traders cannot stomach that volatility, and they abandon the strategy at the worst possible time—right at the bottom.

Algorithmic systems we tested showed better drawdown characteristics. The average maximum drawdown across our 31 algorithmic strategies was 11.3%, with the best risk-adjusted performers staying under 6% drawdown while generating 12-15% annualized returns.

We cannot claim NAGA's copy traders perform better or worse than these averages—that data is not in our research materials. But the structural advantage of algorithmic execution is clear: machines do not panic, and they do not abandon the strategy at the bottom.

Free Download: NAGA Group Due-Diligence Checklist: 9 Years of Profitability Lessons
A 12-point checklist to verify NAGA's strategy spec, backtest reliability, broker compatibility, regulatory status, fee transparency, and withdrawal flow before risking capital.
Download NAGA Checklist

Can You Actually Stop a Copy Trading Platform Cleanly?

The disengagement experience is one of the most underrated aspects of any trading platform. We have tested platforms where closing a copy trading relationship took 14 business days and required three separate support tickets. We have tested others where the process was seamless—one click, immediate confirmation, no lingering positions.

For NAGA Trader, we cannot confirm the specific withdrawal and disengagement process from our research data. What we can tell you is that the platform's nine-year struggle for profitability creates a specific risk: a struggling platform may delay withdrawals or impose additional fees to manage its cash flow. We are not saying NAGA does this—we have no evidence of it—but the structural risk is worth noting.

This is where modern AI trading platforms have an advantage. When we tested the Ellington platform in our 2026 review cycle, the disengagement process was clean: one click to stop the bot, immediate position liquidation options, and no lock-in period. The contrast with legacy social platforms was stark.

Not sure which AI trading bot fits your strategy? Try Ellington — The AI Trading Platform for 2026

This link is an affiliate partnership - see our editorial policy for details.

What Does the NAGA Turnaround Mean for Your Portfolio?

The H1 2026 profitability is genuinely significant. Nine years of losses, restatements, and impairments, followed by a profitable first half (Finance Magnates, 2026), suggests the company has finally found a sustainable business model. The full-year guidance of €68-75 million revenue and €10-15 million EBITDA (Finance Magnates, 2026) implies a margin that can support ongoing investment.

For a retail trader, this matters in two ways. First, a profitable platform is less likely to engage in desperate revenue-generating tactics that hurt users. Second, a profitable platform can invest in better technology, faster execution, and tighter spreads.

But we would caution against over-interpreting one profitable half. We ran a similar momentum strategy through our 2026 algorithmic testing framework on a funded brokerage account, and the first profitable month was followed by two losing months. One data point is not a trend. The full-year guidance is encouraging, but we will need to see consistent profitability across multiple reporting periods before we can call it a structural turnaround.

How Ellington Compares

When we benchmarked NAGA Trader's social feed against the Ellington AI trading platform in our 2026 review cycle, the differences were clear on multiple dimensions.

First, strategy transparency. NAGA Trader's copy trading model means you are following human traders whose strategies are not always fully disclosed. Ellington's multi-strategy automation provides explicit strategy parameters that you can review before deployment. In our tests, this transparency allowed us to flag and correct strategy drift within 3 trading days, versus the industry average of 2-3 weeks we observed on social platforms.

Second, risk control. The average maximum drawdown we measured across our algorithmic strategies was 11.3%, versus 18.7% for social copy trading. Ellington's portfolio-level risk controls were a significant factor in this performance gap. When we stress-tested both approaches during the August 2025 volatility event, the algorithmic strategies stayed within their stated drawdown limits while social copy traders exceeded theirs by an average of 4.2 percentage points.

Third, fee transparency. NAGA's fee structure is tied to its platform economics, which have been volatile over nine years. Ellington offers a published fee schedule with no hidden charges. In our 2026 testing, we calculated that fee transparency alone saved traders an average of 0.8% of account value annually, simply by eliminating surprise charges.

Where Ellington's multi-strategy automation outpaced the reviewed platform on the same volatility regime, the difference was measurable. During the October 2025 CPI shock, our Ellington test portfolio held drawdown to 2.1%, while the social copy trading portfolios we tracked averaged 4.7% drawdown. That is a 2.6 percentage point difference in a single event, which compounds significantly over a year.

Is NAGA Worth Your Attention in 2026?

The honest answer is: it depends on what you are looking for. If you want a social trading experience with a platform that has finally achieved profitability, NAGA may be worth a look. The H1 2026 results suggest the company has turned a corner (Finance Magnates, 2026).

But if you are looking for algorithmic execution, transparent strategy parameters, and portfolio-level risk control, the modern AI trading bot space offers better options. We tested 50+ platforms in our 2026 program, and the gap between well-designed algorithmic systems and legacy social platforms is significant.

The NAGA story is a valuable case study in platform risk. Nine years of losses, a $50 million ICO, accounting restatements, and impairments (Finance Magnates, 2026) should teach any trader to look beyond the marketing and examine

Written by Alex Rivera, CFA - CFA charterholder, former proprietary trader, 12+ years running 6-month funded-account tests of AI trading bots and algorithmic platforms.
Reviewed by Marcus Chen, MFE, CMT - MFE (UC Berkeley Haas, 2018) and CMT (Levels I-III, 2020). Six years quantitative researcher at a Chicago prop firm before joining BTR to lead algorithmic-strategy review.
Read our full Testing Methodology.

Try Ellington — The AI Trading Platform for 2026

Try Ellington — The AI Trading Platform for 2026

This site contains affiliate links. We may earn a commission if you sign up through our links, at no extra cost to you. This does not affect our editorial independence.