iSAM Securities says its broker risk network is now identifying more than 1,500 accounts every month where the same underlying trader appears to be holding materially offsetting positions across different brokerages, highlighting how retail trading surveillance is increasingly moving beyond the boundaries of a single broker.
The figure was disclosed October 5 as iSAM announced four additions to its Radar risk-management platform: Exposure Monitor, Hedge Alerts, LP Reconciliation and Book Switch Impact.
The most significant for retail traders is Hedge Alerts. According to iSAM, Radar uses a proprietary clustering methodology to group related accounts into a single view of the underlying trader, including when those accounts sit with separate brokers participating in the Radar Network.
That allows a participating brokerage to potentially learn that an apparently ordinary customer account forms part of a wider trading strategy elsewhere.
For example, a trader might hold a long position with one broker and an offsetting short position at another. Viewed separately, neither account may look unusual. Radar attempts to identify the relationship between them and determine whether the positions constitute a materially significant hedge.
iSAM says the system is already finding more than 1,500 such cross-broker hedged accounts each month.
Radar Is Moving Broker Surveillance Beyond a Single Customer Database
Traditional broker risk systems largely analyze what happens inside one company: client profitability, position concentration, trade frequency, latency sensitivity and account-level exposure.
Radar Network Alerts goes further.
In its October 5 announcement, iSAM said Hedge Alerts can identify related accounts both within one brokerage and across several firms in its network.
The system measures how much positions within a detected cluster offset one another and applies a minimum gross-position threshold before producing an alert. That threshold is intended to prevent economically trivial or accidental hedges from overwhelming dealing desks.
Radar already offered another network-level feature called Sharp Alerts, which looks for clusters associated with high-frequency trading and adverse broker P&L. Hedge Alerts adds strategies involving materially opposing positions.
iSAM specifically cites swap abuse, bonus abuse and attempts to exploit negative balance protection as examples of behavior brokers may investigate through these alerts.
This development is particularly relevant after recent disputes in which trading firms have relied on technical relationships between accounts. One recent case involved a FundingPips trader whose account was allegedly linked through a device-ID match, illustrating how signals invisible to traders can become important in account-risk decisions.
iSAM Has Not Disclosed What Identifiers Create a Cross-Broker Match
The biggest unanswered question is how Radar determines that accounts held at different companies belong to the same underlying person or trading operation.
iSAM repeatedly describes the process as a “proprietary clustering methodology,” but its public product material does not specify the individual identifiers used to create those clusters.
It therefore remains unclear whether matching is based on customer identity information, technical identifiers, trading behavior, financial data, device or network signals, or some combination of those inputs.
Nor does the October 5 announcement explain exactly what raw customer information participating brokers contribute to the Radar Network, whether matching takes place on pseudonymized identifiers, or how much information about a trader at Broker A becomes visible to Broker B after a match.
iSAM’s general privacy policy says the company uses measures including pseudonymization and encryption where appropriate and allows information sharing in specified circumstances. However, that policy primarily describes iSAM’s relationship with its institutional clients and does not publicly explain the architecture of Radar’s cross-broker clustering system.
That does not imply improper data sharing. It means the public product material currently provides considerably more information about what Radar can discover than about how those discoveries are produced.
Previous Radar Data Shows How Much Broker P&L Is at Stake
The commercial incentive for brokers is straightforward.
iSAM recently said a study using a subset of the Radar network identified 4,180 groups of accounts linked to what it considered the same underlying traders.
In that sample, roughly 1% of logins generated $62 million in broker losses, reducing overall profitability by around 30%. The company also found that 66% of accounts inside flagged groups would not have been individually identified based on high-frequency trading behavior alone.
Separate Radar data covering its wider brokerage base showed that 79.5% of retail clients lost money over a 12-month period, while 20.5% were profitable. Just 1% of profitable accounts reportedly generated 66.5% of total client winnings.
That concentration explains why dealing desks care so much about finding sophisticated traders early.
A broker warehousing customer flow on its own B-book effectively takes the opposite economic exposure to customers. A small group of consistently profitable or highly optimized traders can therefore have a disproportionate effect on broker results.
Recent disputes over alleged latency arbitrage and broker account adjustments demonstrate why the classification of sophisticated trading strategies can become contentious when the broker and customer disagree about whether legitimate trading crossed into prohibited exploitation.
Exposure Monitor Shows Which Individual Logins Are Creating Risk
The second major addition, Exposure Monitor, gives dealing desks a consolidated real-time view of symbol exposure across the broker’s trading platforms.
A brokerage can set warning and breach thresholds for an instrument and receive alerts as its exposure approaches those limits.
The system then allows risk staff to drill down from the overall EUR/USD, gold or index exposure to the individual client logins responsible for the largest positions.
That matters because aggregate exposure by itself does not tell a dealing desk whether risk is broadly distributed among thousands of customers or concentrated in several large accounts.
The difference can determine whether a broker leaves the risk internally, hedges part of it with a liquidity provider or changes the execution profile assigned to particular customers.
Book Switch Impact Puts a Number on A-Book Versus B-Book Decisions
That decision leads directly to another of Radar’s new features.
Book Switch Impact tracks changes in how individual clients are routed and then estimates the financial consequences.
In broad terms, a B-book broker retains customer risk internally. If the customer loses, the broker can benefit economically; if the customer wins, the broker carries that liability.
An A-book arrangement instead passes or hedges the exposure externally with a liquidity provider.
Radar records changes between those execution profiles along with factors such as spreads, markups, time delays and book depletion.
Its new “PnL A” metric then estimates what the broker’s profit or loss might have been if it had retained a customer’s risk instead of hedging it externally.
That gives risk managers a hindsight measurement of the opportunity cost of routing a trader to the A-book.
It could, for example, show that a trader transferred to external hedging because the broker considered the account dangerous later became unprofitable, meaning the brokerage might have made more money by taking the other side internally.
Conversely, it can reinforce an A-book decision if the customer continues generating profits that would otherwise have hit the broker’s own P&L.
The economics of this distinction became particularly visible recently when IG reported weaker OTC revenue retention while Plus500 maintained its outlook. Customer activity alone does not determine a broker’s earnings; how that flow is managed and hedged can materially change the result.
LP Reconciliation Tries to Catch the Risk Created During Book Changes
Changing a client from one execution model to another also creates an operational problem.
If a broker changes how a group of open positions is routed but its liquidity-provider hedge does not update correctly, the firm can unintentionally retain market exposure it believed had been transferred.
Radar’s new LP Reconciliation feature continuously compares exposure across the broker’s trading platforms, bridge and liquidity providers.
When customers are switched between books, it recalculates open exposure and alerts the risk team if those systems no longer reconcile. Radar can then break the difference down by symbol and liquidity provider to show what correcting transaction may be necessary.
The four new tools therefore form a sequence: detect concentrated exposure, identify potentially related traders, change how their risk is handled, and then measure whether that decision worked.
Cross-Broker Intelligence Changes the Balance of Information
The most consequential change is still the network.
A sophisticated trader has historically had one major informational advantage: Broker A generally could not see what the trader was doing at Broker B.
That allowed strategies to be distributed across venues. One account could appear directionally exposed while another neutralized that risk elsewhere.
Network-level clustering begins to erode that separation.
For brokers, this is attractive. It potentially exposes coordinated bonus strategies, swap trades, high-frequency activity and deliberately offset positions that are impossible to identify from one account alone.
For traders, it introduces a different issue: an account may increasingly be assessed not only according to what happens at that brokerage, but according to what a technology provider believes is connected activity elsewhere.
That makes false-positive controls important.
Recent broker account-access disputes show why customers can struggle when internal risk classifications affect their accounts but the underlying evidence is not visible to them.
A cross-broker clustering system magnifies that challenge because some of the evidence may originate outside the broker with which the customer has the direct contractual relationship.
The 1,500-Account Figure Raises Questions Brokers May Eventually Have to Answer
None of this means the 1,500 accounts identified each month are doing anything improper.
Hedging across brokers can have legitimate reasons. Professional and sophisticated traders may deliberately distribute execution, liquidity and counterparty exposure among several venues.
iSAM itself describes Hedge Alerts as tools for identifying relationships and supporting broker investigation rather than proof that every detected cluster represents abuse.
The important question is what happens after the alert.
Does it merely prompt a dealing desk to examine the account? Can it influence A-book or B-book routing automatically? Can a cross-broker match contribute to spread changes, execution controls, account restrictions or termination decisions? And what evidence can a broker provide if a customer challenges an incorrect match?
iSAM’s newest Radar features show just how sophisticated the infrastructure behind a modern retail brokerage has become.
A trader sees an account, a chart and an order ticket.
Behind that interface, a risk system can now measure the trader’s profitability, determine how much market exposure the account creates, compare behavior with other customers, estimate whether the broker should hedge the flow, calculate what a different routing decision would have earned and, increasingly, identify what appears to be the same trader operating at another brokerage.
For brokers, that is powerful risk intelligence.
For retail clients, the more important revelation may be that the walls between separate trading accounts at separate companies are becoming much less opaque than they used to be.
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape.