A-Book vs B-Book: The Broker Execution Models Nobody Talks About Honestly

Introduction: The Industry's Open Secret

If you've traded forex, you've seen the marketing: "True ECN broker!" "100% STP execution!" "No dealing desk!" These claims imply that your broker sends all trades to "the market" and only makes money from commissions or spread markup.

Reality check: The vast majority of retail forex brokers—including regulated ones—operate a B-book, hybrid, or market-making model. They take the other side of at least some client trades. When those clients lose (which is most of the time), the broker profits.

This isn't a scandal. It's just business. But nobody talks about it honestly because "we make money when you lose" isn't great marketing copy.

In this guide, we'll explain A-book, B-book, andhybrid execution models from the broker's perspective. If you're starting a brokerage, this guide will help you choose your model. If you're a trader, this will help you understand what's actually happening to your orders.

Part 1: Definitions (What the Terms Actually Mean)

The forex industry loves buzzwords. Let's define them properly.

A-Book

Definition: The broker routes client orders to external liquidity providers (LPs). The broker only makes money from spread markup or commission.

Also called: STP (Straight-Through Processing), ECN (Electronic Communication Network), NDD (No Dealing Desk)

Reality: Pure A-book brokers are relatively rare because margins are thin and require high volume to be profitable.

B-Book

Definition: The broker internalizes client orders, taking the other side of the trade. When the client loses, the broker profits. When the client wins, the broker pays.

Also called: Market Maker, Dealing Desk, Principal Execution

Reality: Most retail forex volume is B-booked. The broker is betting that clients will lose—and statistically, they're right.

Hybrid

Definition: The broker dynamically routes orders to A-book or B-book based on client profitability profile. Losing clients get B-booked; profitable clients get A-booked.

Also called: Dynamic Execution, Smart Routing, Risk-Based Execution

Reality: This is what most sophisticated brokers actually do. Best of both worlds.

Part 2: A-Book Execution Explained

Let's understand how pure A-book execution actually works.

The Order Flow

1. Client places order in MT5
2. Order hits broker's trading server
3. Liquidity bridge sends order to LP(s)
4. LP executes and confirms fill
5. Bridge confirms back to MT5
6. Client sees executed trade

Total time: 10-100ms depending on infrastructure.

How the Broker Makes Money

• Spread markup: Add 0.3-1.0 pips on top of LP pricing
• Commission: Charge $3-$7 per lot per side
• Swap markup: Add a few points to overnight financing

A-Book Economics Example

That's ~12% margin. Not great. This is why pure A-book is tough—you need serious volume to make it work.

A-Book Pros and Cons

Pros:

• No conflict of interest with clients
• No market risk—you never lose when clients win
• Easier to market as "true ECN"
• Regulators prefer it
• Scales linearly with volume

Cons:

• Lower profit margins
• LP relationship requirements
• Need significant volume (10,000+ lots/month) to be profitable
• Subject to LP issues (fills, rejections, last-look)
• Higher infrastructure complexity

Part 3: B-Book Execution Explained

Now let's talk about the model nobody wants to admit they use.

The Order Flow

1. Client places order in MT5
2. Order hits broker's trading server
3. Broker fills order internally from their own book
4. Client sees executed trade (instant)
5. Broker's position is opposite to client's

Total time: <1ms. Instant execution.

How the Broker Makes Money

• Client losses: When client loses $1,000, broker gains $1,000
• Spread (also): Full spread captured, not just markup
• Swaps (also): Full swap captured

B-Book Economics Example

That's 2-3x more profitable than A-book on the same volume. Now you understand why brokers B-book.

Why B-Book Works Statistically

The foundational fact: 70-90% of retail forex traders lose money. This isn't debatable—regulators publish this data. If you're betting against retail traders as a group, you win in aggregate.

Specific statistics:

• Average retail account loses 30-50% before going dormant
• Average winning trade: 47 pips. Average losing trade: 83 pips (traders hold losers)
• Position bias: Retail is net long 70% of the time (good for broker in ranging/falling markets)

The broker is essentially running a very well-understood casino. The house edge is statistical, and it works.

B-Book Pros and Cons

Pros:

• Much higher profit margins
• No LP dependencies or costs
• Instant execution (better for clients, ironically)
• Lower infrastructure complexity
• Profitable from day one (with any losing clients)

Cons:

• Inherent conflict of interest
• Market risk from winning clients
• Reputation risk if exposed
• Can blow up during extreme volatility
• Harder to scale past certain volume
• Regulatory scrutiny in some jurisdictions

Part 4: The Hybrid Model (What Most Brokers Actually Do)

Smart brokers don't choose A-book OR B-book. They choose BOTH, dynamically.

How Hybrid Works

1. Classify traders: Track P/L, win rate, trade patterns over time
2. Score risk: Assign toxic score (1-100) based on profitability
3. Route accordingly:
   • Score 1-30 (consistent losers): B-book everything
   • Score 30-70 (uncertain): B-book small trades, A-book large
   • Score 70-100 (profitable): A-book immediately
4. Hedge net exposure: If B-book exposure gets too one-sided, hedge in market

Trader Classification Signals

What makes a trader "toxic" (profitable and worth A-booking):

• Win rate >55% over 100+ trades
• Positive expectancy (average win × win rate > average loss × loss rate)
• Short holding times (scalpers often profitable)
• Trades around news (often directionally correct)
• Uses tight stop losses (disciplined risk management)
• Geographic location (certain regions have better traders)

Conversely, signs of "profitable for broker" clients:

• No stop losses (hold and hope)
• Overtrades (high frequency, small accounts)
• Martingale position sizing
• Trades only one direction (always long or always short)
• Makes deposits after stop out (emotional trader)

Hybrid Economics

The beauty of hybrid: you capture B-book profits from the 80% who lose, while hedging exposure from the 20% who might win.

Part 5: The Real Economics Comparison

Let's put all three models side by side with realistic numbers.

This is why almost nobody runs pure A-book. The margin difference is too significant to ignore.

Part 6: Risk Management in B-Book

B-booking isn't just "bet against clients and hope." Sophisticated brokers manage their exposure carefully.

Exposure Monitoring

At any moment, you need to know:

• Net position per symbol: Are clients net long or short EUR/USD?
• Exposure in dollars: What's your P/L if EURUSD moves 100 pips?
• Concentration risk: Is one whale dominating your exposure?

Natural Hedging

If 1,000 clients are long EUR/USD and 800 are short, your net exposure is only 200 lots long. The opposite trades cancel out. This is why B-booking getseasier with more clients—exposures tend to offset.

External Hedging

When net exposure gets too large, hedge in the real market:

• Threshold approach: Hedge when net exposure exceeds $X
• Time-based: Hedge any overnight exposure
• Event-based: Hedge before major news releases

Stop-Loss Orders

If a whale client builds a massive profitable position, the broker may force close at a loss rather than let them run. This is what "stop out manipulation" accusations sometimes refer to—though it's often just the broker protecting their book.

Part 7: Technology Requirements

Different models require different tech stacks.

A-Book Requirements

• Liquidity bridge: FxTrusts Liquidity Bridge or similar
• Multiple LP connections: Aggregated pricing from 3+ providers
• Smart order routing: Route to best price, handle rejections
• Margin management: Track margin with each LP
• Latency optimization: Colocated servers, FIX protocol

Monthly cost: $3,000-$10,000 for bridge + LP fees

B-Book Requirements

• Price feed: From data provider (not LP, since no execution)
• Exposure monitoring: Real-time net position by symbol
• Risk controls: Alerts when exposure exceeds thresholds
• Hedging capability: Ability to hedge in market when needed

Monthly cost: $500-$2,000 for data + monitoring

Hybrid Requirements

• Everything above plus...
• Trader classification engine: Algorithm scoring client profitability
• Dynamic routing: Real-time decision on A-book vs B-book per trade
• Reporting: Track P/L by routing decision to optimize

Part 8: The Regulatory Perspective

How do regulators view A-book vs B-book?

Most Regulators Allow Both

B-booking (market making) is legal in virtually all jurisdictions. Major regulated brokers do it. The key requirements:

• Disclosure: Must explain execution policy in documentation
• Best execution: Must demonstrate fair pricing (no worse than market)
• Capital adequacy: Must hold sufficient capital for risk exposure
• Conflict management: Must have policies to manage conflicts of interest

Regulatory Preferences

While legal, some regulators encourage A-booking:

• CySEC: Strong preference for straight-through processing
• FCA: Heavy disclosure requirements for dealing desk brokers
• ASIC: Increased scrutiny on market-making models

Offshore regulators (St. Vincent, Seychelles) largely don't care which model you use.

Part 9: Choosing Your Model

If you're starting a brokerage, here's how to decide.

Start B-Book If:

• You have limited capital (<$200,000)
• You expect low initial volume
• You need profitability quickly
• Your clients are retail traders (not professionals)
• You're operating offshore

Start A-Book If:

• You're targeting professional/institutional traders
• Regulatory requirements mandate it
• You have significant volume from day one
• "No dealing desk" is your marketing angle
• You have access to cheap liquidity

Start Hybrid If:

• You understand both models
• You have technology for trader classification
• You want to maximize profitability while managing risk
• You're planning to scale

Frequently Asked Questions

Conclusion: The Uncomfortable Truth

Here's the reality of forex brokerage in 2025:

• Most brokers B-book most volume
• They're betting against retail traders
• They're usually right
• This is legal, regulated, and standard practice
• The marketing says otherwise because honesty doesn't sell

Does this make B-book brokers "scams"? No. They provide a legitimate service: access to forex markets for retail traders. The conflict of interest exists, but so does competition—brokers that offer terrible execution or clearly manipulate prices lose clients to better alternatives.

As a broker, choose your model based on business needs, not moral judgment. As a trader, understand what you're dealing with—and trade accordingly.

At FxTrusts, we provide technology for all execution models. Whether you want pure A-book, full B-book, or sophisticated hybrid routing, we have the infrastructure. What you do with it is your business decision.

Just don't pretend to be ECN when you're not. That's the only rule.