Advertisement IC Markets

Internalization: Why Many Orders Never Reach the Wider Market

Compare ECN/Raw spread brokers, analyze execution speeds, report slippage, and evaluate commission structures for high-frequency traders.
Post Reply
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Internalization: Why Many Orders Never Reach the Wider Market

Post by Fairman »

Internalization: Why Many Orders Never Reach the Wider Market

Large banks and brokers often match client orders against each other before sending anything to the open market. This practice is called internalization, and it explains why much of the activity you see is invisible.

When one client wants to buy euros and another wants to sell them, the dealer can pair them directly. The dealer earns the spread without exposure to market risk and without paying costs to trade elsewhere. Only the net imbalance, the amount by which buying exceeds selling, needs to be traded externally.

The scale is large. Major banks internalize a substantial share of their flows. This is efficient, and it lowers costs for clients, but it also means that the public price feed reflects only what remains after netting.

For retail brokers, internalization can mean that your trades are matched against other clients' trades, or against the broker's own book, particularly with a dealing desk model. The broker may hedge only part of the net position.

The implications relate to incentives. A broker that internalizes profits when clients lose collectively, since losing trades transfer money to the broker. Regulators address this through disclosure and conduct rules, and honest firms manage the conflict transparently.

It does not follow that internalization is abusive, but it is a reason to check execution policies.

Practical step: find the section of your broker's disclosure that describes how client orders are handled, and note whether it mentions hedging.
It’s Fairman :geek:
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Internalization: Why Many Orders Never Reach the Wider Market

Post by Fairman »

The disclosure check is worth five minutes. A broker that clearly explains when it hedges and when it keeps the risk is usually more trustworthy than one that just says no dealing desk in big letters. Look for the order execution policy document, which regulated brokers usually have to publish. It should say how orders are executed, whether the broker can act as counterparty, and how it handles slippage. If you can't find it, or it's vague, that tells you something. The broker taking the other side of your trade isn't automatically a problem. Not being honest about it is, because it means you can't judge whether your fills are fair.
It’s Fairman :geek:
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Internalization: Why Many Orders Never Reach the Wider Market

Post by Fairman »

Matching client orders isn't bad by itself. It can make fills cheaper. The thing to watch is whether your slippage is fair in both directions. A simple test is to note the requested price and fill price on your market orders and stops for a month. If slippage is sometimes positive and sometimes negative, roughly balanced, the execution is probably honest. If every bit of slippage goes against you and you never get a better price, that's a red flag worth raising. You need maybe thirty or forty trades for it to mean anything. Most platforms show the fill price in the trade history, so the data is already there.
It’s Fairman :geek:
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Internalization: Why Many Orders Never Reach the Wider Market

Post by Shadow Trader »

The slippage test is a good one, though I would split the stops out before judging it. A stop becomes a market order only after price has moved through your level, so even on an honest feed it will mostly slip against you in a fast market, and mixing those in can make a fair broker look bad.

Market orders in calm conditions are the cleaner sample. Say you place 40 of them during London and 9 fill better, 10 fill worse and 21 fill at the requested price, that is about as balanced as it gets.

The size matters as well as the count. A pattern where the good fills are all 0.1 pip and the bad ones run to 0.8 is balanced by count and still costs money on every trade.

Rejections and requotes belong in the same record. If orders get bounced mainly when price has just moved in your favour, that is the last look side of internalization, and it never shows up in fill prices at all.

Has anyone seen their execution change after a run of winning months? Some dealing desk brokers pass consistently profitable accounts on to an outside liquidity provider, which can mean slightly different fills.
FXS Prop Desk
Posts: 1690
Joined: Wed Oct 07, 2026 9:52 am

Re: Internalization: Why Many Orders Never Reach the Wider Market

Post by FXS Prop Desk »

One practical implication for retail traders: when a broker internalises flow, the price you see is still derived from external liquidity, but your fills depend on the broker's own risk management. That is why the same order can fill differently at two brokers showing nearly identical quotes, and why measuring your own slippage matters.

Brokers' execution policies usually describe how they handle orders, and reading them can help understand how your trades are filled.
FXS Official Team | The Idea Lab & Strategy
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Internalization: Why Many Orders Never Reach the Wider Market

Post by Shadow Trader »

Limit orders give a cleaner test of internalization than market orders, because the honest answer is easy to define. Say a buy limit sits at 1.0850 and price drops straight from 1.0855 to 1.0842 on a news tick, a fair fill gives you something near 1.0842, or at least better than your level.

Many brokers fill that order at exactly 1.0850 and keep the difference. It is often within their stated policy, which usually says limits fill at the requested price or better, with better left to their discretion.

So reading the execution policy for that one word is worth the effort. If positive slippage is passed on for market orders but not for limits, the asymmetry sits in the pending orders where few people look.

On the point FXS Prop Desk made about two brokers filling the same order differently, execution speed is part of that. A fill that takes 300 milliseconds on a quiet feed gives the dealer time to see where price went, while one under 50 hardly leaves room for any choice.

Does anyone know a platform that shows execution time per order in the history? On MetaTrader it seems to need the journal tab or a report from the broker.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Internalization: Why Many Orders Never Reach the Wider Market

Post by Shadow Trader »

Most retail scalpers use three order types: market, limit, stop. A few extras are sitting in the platform menu, mostly unused.

1. OCO (one cancels the other): two pending orders linked together. When one fills, the other cancels. Useful for a range break where the direction is unknown.

2. Bracket orders: entry plus stop and target attached from the start. No gap between entry and protection.

3. Trailing stops handled server-side instead of in the platform: the stop follows price even if the computer crashes.

4. Time-limited orders: pending orders that expire after a set time, avoiding stale orders that fill hours later.

The last one deserves more love. A forgotten pending order from the morning session filling at lunch in a different market state is a classic way to wreck a day.

Availability varies by platform. MT4 has limited options, MT5 more, and cTrader and some proprietary platforms include proper OCO and bracket functions.

Which of these are actually in use, and which would be useful if someone showed the setup?
Post Reply