Weekly Review: AvaTrade and Trade Nation Acquire FXCM; Nasdaq Backs 24/7 Trading

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.

Broker consolidation weeks read like housekeeping news. For anyone running an AI trading bot or an algorithmic trading platform, they are closer to a change of landlord. In a single week the retail forex landscape absorbed two separate deals for pieces of FXCM, a strategic investment from Nasdaq Ventures into a European venue chasing round-the-clock trading, fresh venture funding for a New York prop firm, and a monthly payout disclosure from FTMO that keeps the funded-account economy in the headlines Finance Magnates, Weekly Review.

We benchmarked this week's structural changes against Ellington's multi-strategy automation during our 2026 review cycle, because the question underneath every headline here is the one a bot operator asks before every deployment: who is actually on the other side of my order, and what changes when the plumbing moves?

What actually happened in the FXCM deals?

AvaTrade has agreed to acquire the majority of FXCM Group's business and brand rights, bringing two established CFD brokers under the same corporate group. Financial terms were not disclosed. Finance Magnates had reported in June that AvaTrade was seeking to acquire Stratos, the parent of FXCM and Tradu, from investment bank Jefferies, which had controlled Stratos for a decade after providing a liquidity rescue following the 2015 Swiss franc crisis Finance Magnates.

Note the structure. This was not a single acquisition of a whole business. AvaTrade is taking the majority of the remaining operations, which leaves a residue of the estate in other hands.

The UK client book went to a different buyer

Days earlier, Trade Nation acquired FXCM's UK client book for an undisclosed sum, a deal Finance Magnates confirmed on Tuesday. Matthew Wright, Trade Nation's chief strategy officer and UK CEO, cited the longevity of FXCM's client base as a factor that attracted the company to the portfolio Finance Magnates.

So the brand went one way and a regulated client population went another. For a discretionary trader that is an administrative event. For someone with an API key, a rented VPS and a strategy running on live capital, it is a change of counterparty, contract and potentially execution venue, all landing at once.

Every financial term in this table is undisclosed in the source reporting. That matters for our readers because it means the client-facing terms, the ones that actually determine whether a bot keeps working, are also likely to be negotiated quietly.

Why the split matters more than the price

We have flagged this pattern before in our review cycle, and we will flag it again: the value of a broker to an algorithmic trader is concentrated in three things that never appear on a marketing page, which are fill logic, API rate limits and the legal entity holding the account. Client-book transfers touch all three. Nothing in the public reporting tells us how FXCM's UK book will be re-papered, and we will not speculate. What we can say is that any trader running automation through a transferring book should treat the migration as a re-deployment, not a rebrand.

Which broker will your bot be executing against in 2027?

The quiet infrastructure story this week is more consequential for automated strategies than the M&A headline. Axi has migrated the majority of its clients to MetaTrader 5, with more than 60 percent of customers now using the platform according to a case study published by MetaQuotes. The broker has also begun deploying Ultency, MetaQuotes' order-matching engine for MT5 brokers, having previously used a competing order-management system from Your Bourse that it adopted in 2025 Finance Magnates.

MetaQuotes noted that earlier Ultency users were largely liquidity providers, which makes Axi's adoption an extension of the engine into a retail brokerage operation rather than a like-for-like replacement. Translation for a bot operator: the code path your orders travel has changed, and the party that built your matching engine is now also the party that builds your terminal.

Free Download: FXCM Broker Due-Diligence Checklist for AI Trading Bots After AvaTrade/Trade Nation Acquisition
Use this checklist to verify broker compatibility, regulatory status, fee transparency, and withdrawal flow for AI bots trading FXCM under new ownership and Nasdaq's 24/7 market push.
Get FXCM Bot Checklist

The Trading 212 numbers are worth pausing on. A group that invested 44 million pounds across overseas subsidiaries, with 21.1 million pounds directed to Germany and roughly 4 million euros paid to acquire the German entity it now runs as a licensed bank operation generating 2.7 million pounds of revenue, is a group optimising for jurisdiction-by-jurisdiction licensing Finance Magnates. For an algo trader, jurisdiction is not a detail. It determines leverage caps, hedging rules and whether your strategy is even legal in the form you wrote it.

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.

Does 24/7 trading help or hurt automated strategies?

Nasdaq Ventures has made an undisclosed strategic investment in One Trading, supporting discussions on round-the-clock trading infrastructure and potential joint initiatives. The partnership will explore extending One Trading's continuous trading capabilities into traditional derivatives. Both companies describe the arrangement as assessing areas of cooperation rather than a completed integration Finance Magnates.

The demand signal is real. CME Group reported that its first six weekends of cryptocurrency futures and options trading generated 1 billion dollars in volume, while average daily volume in crypto derivatives rose 44 percent year on year in the first half to 280,000 contracts. SEC Commissioner Hester Peirce has observed that many market participants view extended trading as inevitable if unwelcome, citing thinner order books, wider spreads and increased price volatility Finance Magnates.

Here is the part of the 24/7 debate that automated strategy builders are underweighting. When a venue moves to continuous trading, it does not simply add hours to the end of the week. It changes which liquidity cohort is present at any given moment, and it changes the clock against which a strategy's statistics were computed. A momentum model calibrated on a five-session week, with an annualisation factor built around roughly 252 sessions, does not carry those same statistics into a seven-day market honestly. Composite a weekend tape into the same series and your realised volatility estimate, your Sharpe ratio and your expected maximum drawdown all shift, not because the strategy changed, but because the sample definition did.

We re-implemented that clock problem inside our backtest harness using the published CME volumes as the sizing reference, and the distortion is structural rather than cosmetic. Any bot that infers session from a server timestamp will mislabel weekend bars, and any bot whose edge is tied to a specific liquidity window, say the London or New York open, will find that edge diluted or relocated once a venue trades continuously. The failure mode is not a blowup, it is a slow drift in a metric you thought was stable.

There is a second-order risk here too. CME has publicly wound down its plan for 24/7 markets over weekend staffing concerns, which is a useful reminder that the operational overhead of continuous markets falls on venues and brokers before it falls on traders. If your broker is running extended hours on a skeleton crew, the counterparty risk profile of your weekend exposure is not the same as your Tuesday afternoon exposure. That is a portfolio-level concern, not a strategy-level one, and it does not get modelled in most retail backtests.

What the prop firm payout numbers really say

FTMO distributed more than 14.4 million dollars in rewards to approximately 7,100 traders in September, taking its reported cumulative payouts above 650 million dollars. The Prague-based firm said it paid more than 200 million dollars over the preceding twelve months, an average of around 16 million dollars per month. September's rewards were roughly 5 percent lower than August's level, and the implied average payment was about 2,030 dollars. FTMO also reports more than 4.5 million registered traders worldwide Finance Magnates.

Hola Prime reported paying more than 10 million dollars to funded traders, more than tripling its cumulative total since April. The company put the average payout at approximately 1,800 dollars, down from around 4,500 dollars when it reported distributing about 3.2 million dollars earlier in the year. A Deloitte review earlier in 2026 examined how quickly payouts were processed through mid-March but did not verify the subsequent cumulative total Finance Magnates.

The FTMO and Hola Prime figures illustrate a contrast that matters when you are deciding where to run an automated strategy under a funded model. FTMO reports a high-frequency monthly cadence with a stable implied average payment; Hola Prime's cumulative has grown fast while its average payment has fallen from roughly 4,500 dollars to about 1,800 dollars. The same total can be produced by more traders earning less, which changes the economics of running a high-turnover bot on that infrastructure.

Vest Labs is a different animal. A 13 million dollar pre-seed is unusual in a sector where most firms self-fund, and the angel list is a signal about how institutional adjacent capital views the prop model Finance Magnates. It tells us nothing about payout reliability, which is the only prop firm metric that matters to a bot operator.

Can you actually switch the bot off cleanly?

This is the disengagement question, and it is the one we test hardest in our live-trading evaluation framework. In a funded-account context, clean disengagement means you can flatten open positions, revoke API credentials and receive whatever balance is owed without triggering a rule violation you did not know existed. In a broker client-book transfer, it means your automation permissions, your VPS whitelisting and your historical trade data survive the move. We have not run a verified test through a completed client-book migration, and we will not publish a disengagement timing figure we cannot source. Treat this as an open item to confirm in writing with both the transferring and receiving entity.

Is your bot provider on a regulated register?

This is where we stay deliberately strict. Nothing in the source material for this week's deals includes a licence number for AvaTrade, Trade Nation, FXCM, One Trading, FTMO, Hola Prime or Vest Labs, and we do not assert regulatory status we cannot cite. If you are about to route automation through a broker involved in a transfer, check the primary register entry yourself rather than a marketing page. In the UK that means the FCA register search. In Australia it means the ASIC Connect registers.

The distinction that trip people up is between the broker's permission and the bot vendor's status. An AI signal provider may be entirely unregulated while your broker is fully authorised, and that asymmetry is where retail money gets hurt. Our standing rule is simple: if a vendor's regulatory claim cannot be traced to a primary register entry, treat it as unverified regardless of how the landing page is worded.

How we would test a strategy through a broker migration

Our 2026 algorithmic testing program treats any change to the matching engine or the account-holding entity as a re-certification event. We would re-run the strategy specification against the new venue, log every deviation between intended and executed behaviour, and compare the backtest versus live-trade performance gap before and after the change. We are not publishing a deviation count for a migration we have not personally completed, and we would rather say so than manufacture one.

What we can do is name the checkpoints. First, confirm that the symbol universe and contract specifications are unchanged, because a renamed instrument can silently break a bot that keys off tickers. Second, confirm that order types behave identically, since a matching engine change can alter how stop orders are triggered in thin conditions. Third, confirm that API credentials and permissions transfer, because a revoked key is a flat position you did not choose. The Axi transition is the live example of this in the current news cycle, with a MetaQuotes order-matching engine moving into a retail brokerage setting where it had previously been used largely by liquidity providers.

How Ellington compares on broker-agnostic execution

This is the dimension where the week's news exposes a structural weakness in how most retail automation is deployed. When a single broker's client book can be split between two buyers, or a single order-matching engine can be swapped underneath your strategy, the execution assumption at the base of your stack is doing more work than your signal logic. Where Ellington's multi-strategy automation stood out across our review period was in holding portfolio-level risk sizing across strategies rather than inheriting the constraints of one venue's paper. On the same volatility regime, a book that relies on a single API integration inherits migration risk that a broker-agnostic, multi-strategy engine does not.

That is not an argument that any platform is immune to counterparty change. It is an argument that concentration in one execution path is an under-priced risk, and this week gave us a clean example of why.

Not sure which AI trading bot fits your strategy? See how Ellington handles portfolio-level risk across venues

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

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.

Frequently Asked Questions

Does an AI trading bot keep working after my broker sells its client book?

Sometimes, but you should assume nothing until you have it in writing. Client-book transfers can change the legal entity, the platform, the leverage schedule and API permissions, and none of the FXCM-related deals disclosed their financial or client-facing terms. Re-verify API access and symbol mappings with the receiving entity before the transfer date.

Will 24/7 trading change how my algorithm's drawdown is measured?

Yes, if you are not careful with your clock. A strategy annualised on roughly 252 weekday sessions produces different volatility, Sharpe and expected maximum drawdown figures when weekend bars are added to the series. The strategy has not changed; the sample definition has.

Can I run an AI trading bot on a prop firm account like FTMO or Hola Prime?

Most funded-account programmes permit automation within their rule sets, but rule engines typically restrict strategies that run during news events or hold through weekend gaps. FTMO reported paying approximately 7,100 traders in September with an average payment of about 2,030 dollars. Hola Prime's reported average payout was about 1,800 dollars. Confirm automation permissions and consistency rules in writing before deploying.

What happens if the API connection drops mid-trade?

The position stays open at the broker, only your ability to manage it stops. Server-side stop and take-profit orders placed with the broker survive a dropped connection; orders your bot manages locally do not. This is the single most common way a well-backtested strategy produces a live result it never showed in testing.

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

We have no verified data on US availability for the platforms in this week's news, and we will not guess. Pattern Day Trader rules apply to margin equity accounts and cap day trades below 25,000 dollars in equity, which can make high-frequency automation structurally unusable for smaller US accounts regardless of how the strategy performs.

How do I check whether a broker or bot provider is properly regulated?

Use the primary register, not the marketing page. The FCA register search and the ASIC Connect registers are the starting points for UK and Australian permissions. If a vendor's claim cannot be traced to a primary register entry, treat it as unverified.

Are backtests still reliable once a broker changes its order-matching engine?

Backtest reliability degrades whenever fill assumptions change, and a new matching engine is a fill-assumption change. Axi's Ultency rollout, with more than 60 percent of clients now on MetaTrader 5, is exactly the scenario where a previously validated backtest should be re-certified rather than trusted.

How much should I expect to pay in bot subscription fees?

The source material for this week's news does not include subscription pricing for any of the platforms discussed, so we are not publishing a number. What we can say is that any recurring fee is a direct drag on strategy economics and should be modelled against your expected gross edge before deployment, not after.

Why does the FXCM split matter if I do not trade FXCM?

Because it demonstrates how quickly the execution layer under a retail strategy can be restructured. Two buyers, an undisclosed price, and a decade-long ownership hangover from a 2015 liquidity rescue. If your automation assumes one stable counterparty, this week is the argument against that assumption.

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: Trading Industry News.