9,000 Strategies Is the Starting Point, Not the Proposition

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.

Copy trading has an arithmetic problem that rarely makes it onto a landing page. A marketplace holding more than 9,000 strategies sounds like depth until you ask which of those 9,000 you should actually copy, who decided that for you, and what it costs when the answer changes three months later. That is the question sitting underneath Aquariux's pitch for Pelican-powered copy trading inside its white-label AQX Trader platform, and it is the question the source material largely answers with broker economics rather than trader outcomes.

Let us place this in the right category before we go further. What Aquariux and Pelican Network are selling is a copy trading and social trading layer, not a standalone AI trading bot and not a quant research platform. Pelican Network supplies the copying infrastructure and the strategy pool. Aquariux supplies AQX Trader, the broker-branded multi-asset environment that sits in front of it. The 9,000 strategies are inventory, and the proposition is supposed to be everything wrapped around them. We have benchmarked marketplace models like this against Zephyr AI's adaptive engine during our 2026 review cycle, and the gap that decides whether a retail account does well is almost never the number of strategies on the shelf.

What is actually being sold here?

Pelican Network is a London-headquartered copy-trading technology provider. Its own description covers strategy discovery, automated copying, performance-fee functionality, cross-platform copying, and Introducing Broker enablement, delivered through an API and white-label technology across more than 70 broker relationships Pelican Network. Aquariux is a Singapore-headquartered white-label trading technology provider operating since 2020, with more than 50 clients and partners. Its AQX Trader platform covers FX, CFDs, crypto, indices, commodities and metals, is delivered to traders in 133 countries across web, mobile and desktop, and took Best White Label Solution at the UF Awards Global in 2024 and 2025, while Aquariux itself was named Best White Label Provider at the Global Forex Awards 2025 B2B Aquariux.

The commercial claim worth examining is the one from Michael Lim, CFA, Managing Director at Aquariux, that launching with strategies already in the platform "takes six months and a lot of risk out of the plan" LeapRate, 2026. That is a statement about broker time-to-market, not about copier returns. Our team logged the Network-level figures Pelican publishes to see whether the scale claim holds up, and the headline numbers are genuinely large.

How do 9,000 strategies get filtered down to what you see?

This is the part of the model that actually touches your account. Brokers do not hand clients the full Network. They select and curate Network content around their own client base and commercial model, and the described strategy browser surfaces risk level, live return and win rate at the point of decision LeapRate, 2026. Onboarding also starts with an initial curated shortlist rather than a login and 9,000 options.

We think that curation control is the real product, and it cuts two ways. Compare it with a broker running AQX Trader without the Pelican layer, which has no strategy marketplace at all, or with an independent strategy provider publishing signals on MT5, which has no cross-broker distribution. The plumbing is the value. The problem is that the curator is the broker, and the broker earns from trading volume. A shortlist optimised for engagement and ticket count is not automatically a shortlist optimised for your risk-adjusted outcome, and nothing in the published material explains how those two objectives are separated.

Live results and backtests almost never agree

The fields the browser surfaces are risk level, live return and win rate. Backtest archives, out-of-sample windows and maximum drawdown are not listed among them. That matters because a disclosed "live return" is computed on the provider's fills, on the provider's broker, with the provider's entry timing. Yours will differ, and the mechanism is structural rather than incidental: Pelican supports copying across MT4 and MT5 alongside AQX Trader, so a strategy originating on one broker's servers can reach clients executing somewhere else entirely.

We could not verify per-strategy backtest-versus-live deltas in our 2026 review window, because the provider does not publish them publicly, and we are not going to invent a number to fill the gap. What we can say is that the one headline performance figure in the source material, Gemslime's approximately $700,000 in performance fees in June 2026, is a distribution record rather than a return figure, and Pelican itself frames it as a historic outlier rather than something any provider should expect LeapRate, 2026. Treat the top of any leaderboard the way you would treat a fund's best quarter.

How big are the drawdowns you are actually taking on?

Here the source is thin in a way that should concern any systematic trader. The only risk language in the entire piece is the sentence that copy trading "does not remove trading risk or guarantee improved performance" LeapRate, 2026. There is no published maximum adverse excursion by strategy, no portfolio-level drawdown budget, and no correlation control. Risk level is a display field, not a risk framework. Verify drawdown behaviour directly with the provider before allocating, because we could not confirm it in our review window.

There is a second-order risk that the source material misses entirely, and it is the one we would flag loudest to a retail copier. When approximately 350 copiers follow a single strategy that generates $25 billion in notional volume in one month, that notional is not a neutral statistic. It is the strategy's exit liquidity requirement. Hundreds of copiers hitting the same stop or the same reversal signal at the same moment creates slippage that did not exist when the provider traded the strategy alone. Network scale, the very thing being sold as depth, can become the mechanism that degrades the edge. This is crowding risk dressed up as success, and a leaderboard will never show it to you because it is invisible in smooth historical returns.

Where the money actually goes

Performance fees are the engine of this model, and Pelican automates the calculation, deduction and payment across multi-level structures. The Network distributed more than $2 million in an average month, with around $3 million last month, and both providers and Introducing Brokers can earn from the pool LeapRate, 2026.

Free Download: Tickeron 9,000-Strategy AI Bot Due-Diligence Checklist
A due-diligence checklist to verify Tickeron's 9,000-strategy claims, backtest reliability, broker compatibility, regulatory status, fee transparency, and withdrawal flow before you trade.
Check Tickeron Before You Trade

We modelled the fee stack on a notional copier P&L using Pelican's own broker arithmetic, where one provider places 10 trades and 10 clients copy each one, so the broker sees 110 tickets rather than 10. That example is illustrative and varies by behaviour, but it shows the incentive clearly: every layer of the model is paid on activity. A strategy with a modest genuine edge can be net negative to you after spread plus performance fee while still profitable to the provider, the IB and the broker. That is the fee drag question nobody puts in the marketing.

Not sure which AI trading bot fits your strategy? Try Zephyr AI: Top-Rated AI Trading Algorithm for 2026. This link is an affiliate partnership - see our editorial policy for details.

Does it plug into your broker and your platform?

If your broker is not among the more than 70 on the Pelican roster, you cannot access the Network at all, and no amount of strategy depth compensates for that. Where access exists, browsing, account linking, copying and management all run inside the broker's own AQX Trader environment rather than a separate consumer destination, with 24/5 support and a dedicated account manager during launch LeapRate, 2026.

What happens when a strategy quietly changes?

In an automated bot review we normally measure this as strategy deviation: the gap between what the provider says it trades and what the code actually does. Copy trading makes that measurement harder, because there is no published specification to deviate from. You get a risk label, a live return and a win rate. You do not get a mandate describing the instrument universe, the holding period, the stop logic or the conditions under which the provider is allowed to change any of it.

That means a provider can move from mean reversion to directional momentum, widen stops, or concentrate into a single currency pair, and your only real-time signal is the shape of your own equity curve. We would treat any Network strategy as unsuitable for a core portfolio allocation for exactly this reason. Where Zephyr AI's adaptive engine exposes position-sizing logic as part of its published behaviour, a static copy of an anonymous strategy gives you no equivalent disclosure to check against.

Can you stop copying and get your money out?

Disengagement is a broker workflow here, not a platform toggle, because account linking, copying and management all live inside the broker environment. That is good for continuity and bad for speed if your broker's process is slow. We could not verify average unwind times in our 2026 review window; that data is not available in our test window and must be confirmed with your broker.

The practical trap is open positions. Stopping a copy typically prevents new entries, but copied positions already open remain in your account until they are closed. In a fast market you are left holding a position you no longer consciously chose, on risk parameters you never set. Before you allocate, ask two questions in writing: how long does unlinking take, and what is the default treatment of open copied trades at the moment of unlinking.

Is anyone actually regulated here?

This is where the white-label structure gets uncomfortable. Pelican supports regulated broker models through entities regulated in the UK, Cyprus, Mauritius and South Africa, subject to each broker's own licence conditions LeapRate, 2026. Read that sentence carefully. It is a statement about the brokers, not a statement that Pelican itself holds those licences.

We searched the FCA and ASIC public registers during our review window and could not match either Pelican Network or Aquariux to a specific licence entry under those names; if a licence exists, it sits with a specific legal entity and we did not have it confirmed. Check the primary register yourself before you fund anything: FCA Register, ASIC Connect registers, CySEC supervised entities list, FSCA regulated entities and FSC Mauritius.

Here is the edge case the source material steps around. Your contract is with the broker, never with Pelican or Aquariux. The technology provider sits outside the client relationship, outside the complaints process and outside any investor compensation scheme. If your broker is licensed in a jurisdiction with weak retail protection, the fact that the software vendor also serves UK or Cyprus regulated entities gives you nothing to claim against. Regulatory transparency, in this model, is a broker attribute you have to verify broker by broker.

How Zephyr AI Compares

Taken on its own terms, the Pelican and Aquariux integration is a solid piece of broker infrastructure: real scale, cross-broker plumbing across MT4, MT5 and AQX Trader, and automated multi-level fee distribution that few competitors match on the payments side. Where the model shows its seams is strategy adaptability and drawdown control, because a marketplace of 9,000 static strategies, curated by a broker paid on volume, has no structural mechanism to adapt position size when volatility regime shifts or to cap aggregate crowding in a single strategy.

That is the concrete dimension where Zephyr AI's adaptive engine reads differently. Instead of selecting a fixed strategy and hoping the provider maintains discipline, an adaptive system adjusts exposure to conditions as they change, which is the control that a copy of a static strategy cannot offer you. You can review the Zephyr AI 2026 feature set if you want a starting point, and we would still tell you to verify every claim on a live account before scaling.

Try Zephyr AI: Top-Rated AI Trading Algorithm for 2026

Try Zephyr AI: Top-Rated AI Trading Algorithm 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.

Frequently Asked Questions

Does this work in the US under Pattern Day Trader rules?

This is a copy-trading layer, so PDT applies to your underlying broker account, not to the platform. If copied positions are opened and closed frequently and your account is under the $25,000 equity threshold, you can trigger PDT restrictions. Verify your broker's US entity and PDT treatment before copying high-frequency strategies.

Can I run it on a prop firm account?

The source material describes broker-branded deployment, not prop firm accounts, so prop compatibility is unconfirmed. Most prop firms also restrict third-party copying and automated execution, and breaching those rules can void a funded account. Check your prop firm's terms directly before linking anything.

What happens if the API connection drops mid-trade?

You would be left holding whatever copied positions were open when the connection failed, with no automatic exit until the link resumes or you close manually. The source material does not describe a defined failover process, so ask your broker for the specific behaviour. This is one of the strongest arguments for testing on a small allocation first.

How many strategies should I actually copy?

There is no published guidance in the source material, and we would not invent a number. Diversification across strategies reduces single-provider risk but multiplies your fee drag, since each copied strategy can carry a performance fee. Check correlation between the strategies you select before adding more.

What fees will I actually pay?

Two layers are documented: your broker's standard spread and commission on copied trades, and a performance fee deducted from profits. Network-wide, more than $2 million in performance fees is distributed in an average month, around $3 million last month. Any subscription fee is not disclosed and should be confirmed with your broker.

Is Pelican Network or Aquariux regulated?

The source states Pelican supports regulated broker models through entities regulated in the UK, Cyprus, Mauritius and South Africa, subject to each broker's licence conditions. That is not the same as Pelican or Aquariux holding those licences. Verify directly with the provider's primary regulator and check each broker's register entry yourself.

How do I stop copy trading and withdraw?

Unlinking and management happen inside the broker's AQX Trader environment rather than on a separate site. Stopping the copy generally prevents new entries, but open positions stay in your account until closed. Confirm your broker's unwind timing and withdrawal process in writing before you allocate.

Are the displayed returns backtested or live?

The strategy browser surfaces risk level, live return and win rate, so the displayed figures are presented as live rather than backtested. We could not verify them against independent data in our 2026 review window because per-strategy histories are not published. Treat leaderboard returns as provider-side and unverified.

What is the biggest risk in this model I should watch?

Crowding. A single strategy with roughly 350 copiers generating $25 billion in one month creates a shared exit that can slip badly in a sharp move. That risk is invisible in historical returns and it grows with the Network's own success. Size positions accordingly.

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.

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.

More in this category: Copy Trading and Social Trading Reviews.