BEGINNER ARTICLE 4 OF 5

Forex Broker Execution Models: Market Maker, A-Book, B-Book, STP and ECN

8 min read

Lesson purpose: Understand common broker execution labels, the conflicts they can create and the evidence required to evaluate actual fill quality.

The opening scene

One broker advertises “true ECN.” Another says “no dealing desk.” A third openly describes itself as a market maker. Beginners are often taught that one label is automatically good and another automatically bad.

Real execution is more complicated. A broker can use several methods at once, change hedging by product or client flow, and still be required to follow its legal execution policy. The label is the beginning of due diligence, not the conclusion.

A strong explanation of forex broker execution models should connect the visible trading screen with the hidden mechanics underneath it. That includes the product specification, legal entity, data source, price convention, transaction cost and risk limit. The goal of this lesson is not to make a beginner feel certain. It is to make the beginner more precise.

What you will learn

How forex broker execution models works in practical terms.
Which details are controlled by the market and which are controlled by a broker or platform.
How to calculate, verify or document the important numbers.
What professional market participants consider that beginners often miss.
How to avoid turning an educational idea into an untested trade signal.

The broker as contractual counterparty

In many retail OTC forex and CFD accounts, the broker or product issuer is the client’s legal counterparty. The client’s profit can be the provider’s liability, even when the provider later hedges the risk elsewhere.

This creates potential conflicts, but the existence of a conflict is not proof of misconduct. Regulation, governance, pricing controls, capital and complaint processes are designed to manage conflicts.

The practical rule is to document the definition before using it. When the topic depends on a platform, broker or legal entity, record that information beside the number. This prevents a valid statement about one account from becoming a false statement about the entire forex market.

A-Book and external hedging

A-Book commonly describes client exposure that the broker offsets with an external liquidity provider. The hedge can be trade-by-trade, aggregated or triggered when risk limits are reached.

It does not necessarily mean the client’s exact order is passed unchanged to an exchange. Most spot FX is OTC, and the broker can remain the contractual counterparty while managing its market risk externally.

B-Book and internalisation

B-Book generally means the broker retains some client market exposure rather than immediately hedging it. Internalisation can reduce external transaction costs and allow the broker to match opposing client flow.

The conflict is clear: client losses can improve the broker’s result before hedging and operating costs. That makes fair pricing, independent oversight and transparent execution data important. B-Book is not automatically fraud, just as A-Book is not automatically fair.

STP and ECN language

Straight-through processing can describe automated routing without manual intervention. ECN can describe an electronic network displaying or matching multiple participants. Retail marketing uses both terms inconsistently.

Useful questions include: Who is the legal counterparty? Is displayed depth firm? Can orders receive partial fills? Are commissions charged? Does the provider use last look? What happens when liquidity is unavailable?

Hybrid risk management

Many brokers use hybrid models. Small client positions may be internalised while net exposure above a threshold is hedged. Profitable, high-volume or event-sensitive flow can be treated differently from stable retail flow.

A client usually cannot observe the broker’s internal risk classification. Evaluation should therefore focus on measurable outcomes and disclosed policy rather than guessing which book contains the order.

Evidence of execution quality

Relevant evidence includes average and tail slippage, positive versus negative slippage, rejection rate, price improvement, fill speed, spread distribution and complaints. The results should be separated by order type, size, session and event conditions.

A fast average fill can still be poor if the price is consistently worse. One rejected trade can be normal; a repeated asymmetric pattern deserves investigation with complete records.

Finance Chronicles insight

Better broker questions

Replace “Are you ECN?” with:

Which legal entity is the counterparty?
How is the reference price constructed?
Is execution principal, agency, matched principal or another model?
How are conflicts managed?
What is the slippage distribution?
Are positive and negative slippage treated symmetrically?
What are reject and requote rates?
Which orders are server-side?
Is there an independent dispute-resolution body?

This evidence-first approach is more useful than a badge.

This section adds context that is often absent from introductory courses. It does not make the topic more complicated for the sake of complexity. It shows where a simple rule can fail when it meets real execution, legal or statistical conditions.

How an institutional desk sees it

Institutional execution teams evaluate venue quality by size, price, fill probability, market impact, information leakage and credit. They may route different portions of an order to different venues.

Retail traders can perform a smaller transaction-cost analysis. Save every requested price and fill, then compare execution by session and order type. A broker model should be judged by consistent outcomes and legal protections, not by one marketing phrase.

A beginner does not need institutional technology or capital to adopt institutional discipline. The transferable habits are defining exposure, measuring costs, separating facts from interpretation, keeping records and deciding the maximum acceptable loss before taking risk.

Worked example

A trader records 200 market orders:

Positive slippage: 55 trades, average +0.12 pip
No measurable slippage: 65 trades
Negative slippage: 80 trades, average −0.28 pip
Ten worst news trades: average −2.4 pips
Average fill time: 90 milliseconds

The 90-millisecond statistic sounds impressive, but the distribution shows larger adverse than positive slippage. The next step is to compare order size, event timing and reference quotes—not to conclude immediately that the broker is dishonest or excellent.

Verification steps

Identify the exact currency pair, product and legal account type.
Write every input before performing the calculation.
State whether the figure is advertised, observed, estimated or independently tested.
Add spread, commission, financing, conversion and possible slippage where relevant.
Express the result in account currency and as a percentage of equity.
Write what evidence would invalidate the conclusion.

Myth versus reality

Myth: A-Book means no conflict of interest.

Reality: The broker can still be the counterparty and choose liquidity, markup and execution rules.

Myth: B-Book means the broker manipulates every trade.

Reality: Internalisation creates a conflict but is not proof of manipulation.

Myth: ECN is a protected legal category.

Reality: Retail use of the label varies.

Myth: The fastest execution is always the best execution.

Reality: Price quality, fill rate and slippage also matter.

Common beginner mistakes

Judging execution from one trade: A distribution across many comparable trades is needed.
Ignoring positive slippage: A fair study measures both directions.
Mixing market and limit orders: Their expected fill behaviour differs.
Assuming group regulation covers the account: Execution obligations belong to the contractual entity.

Practical exercise

Build a demo execution-quality worksheet for 100 orders. Record:

Order type
Pair and size
Bid and ask at submission
Midpoint benchmark
Fill price
Slippage direction and size
Fill time
Session and event status
Reject or requote
Broker entity

Calculate average, median, best, worst and 90th-percentile adverse slippage. The exercise is about methodology, not publishing a broker verdict from a small sample.

Complete the exercise in a demo environment, spreadsheet or journal. No live position is required. The objective is to practise a repeatable method and identify missing information before money is exposed.

Five-question knowledge check

Who is often the counterparty in retail OTC trading?
Does A-Book prove exchange execution?
What is internalisation?
Is ECN usage consistent across retail brokers?
What data are more useful than a label?

Show answers

1. The broker or product issuer named in the agreement.

2. No.

3. The broker retains or matches some client exposure internally.

4. No.

5. Slippage, fills, rejects, costs and the execution policy.

Final takeaway

Understanding forex broker execution models means more than recognising a definition. The reader should be able to explain the mechanism, identify the variables controlled by the broker or venue, calculate the financial effect and state the remaining uncertainty. That standard is more useful than memorising a rule without knowing when it stops working.

Related lessons

Previous lesson: Beginner Forex Indicators
Next lesson: How to Verify Forex Broker Regulation and Licences

Authoritative sources

Editorial and risk disclosure

This lesson is provided for educational and informational purposes only. It does not constitute financial, investment, legal, tax or trading advice. Forex, CFDs and other leveraged products involve substantial risk. Product rules, leverage, client protections and legal availability differ by jurisdiction, legal entity, client classification and platform.

Finance Chronicles Education Desk · Reviewed 2026-07-10