Market Microstructure

Every time you click "buy" or "sell," your order goes on a journey. It leaves your trading platform, hits your broker's server, gets matched against available liquidity, and comes back as a fill. That entire process — from click to fill — takes milliseconds, but the details of how it works profoundly affect your trading costs, execution quality, and ultimately your profitability. Most traders never think about it. Professionals obsess over it.

How Forex Execution Actually Works

Unlike stock exchanges where all orders go to a single venue, forex is decentralized. Your order can be routed to any number of venues depending on your broker's setup. Understanding the execution chain is step one.

When you place a trade with a retail broker, one of two things happens:

A-Book (STP/ECN): Your order gets passed through to the broker's liquidity providers — banks, non-bank market makers, ECNs. The broker acts as an intermediary, earning money from commissions or a small markup on the spread. They don't care whether you win or lose because they've already hedged your order. This is the ideal scenario for the trader.

B-Book (Market Making): The broker takes the other side of your trade internally. They don't pass it to an external liquidity provider. If you buy, they sell to you from their own book. If you lose (and statistically, most retail traders do), the broker pockets your loss as profit. If you win, they pay out of pocket. Most retail brokers B-book some or all of their flow, especially from smaller accounts.

The uncomfortable truth: B-booking isn't inherently evil. It actually provides tighter spreads and faster execution for retail traders (no need to find external liquidity). The problem arises when brokers engage in unethical practices — widening spreads during your trades, delaying execution, or rejecting profitable orders. Well-regulated B-book brokers operate fairly because regulators audit their practices. Unregulated ones can manipulate freely.

Market Makers vs ECN/STP

Market makers provide liquidity by quoting both a bid and ask price. They profit from the spread — buying at the bid and selling at the ask. In forex, market makers range from tier-1 banks (JP Morgan, Citi) to retail brokers running dealing desks.

Key characteristics of market maker execution:

• They set their own bid/ask prices
• Fixed or stable spreads (they absorb volatility risk)
• Guaranteed fills in normal conditions (they're obligated to take the other side)
• Potential conflict of interest if the broker is your counterparty
• During extreme volatility, they may widen spreads dramatically or reject orders

ECN (Electronic Communication Network) brokers aggregate quotes from multiple liquidity providers and show you the best bid and ask from the entire pool. Your order matches against the best available price from any connected provider.

Key characteristics of ECN execution:

• Variable spreads based on actual market liquidity
• Can be extremely tight (0.0-0.3 pips on EUR/USD) during peak hours
• Can blow out during news events or low-liquidity periods
• Usually charges a commission per lot traded ($3-7 per side)
• No dealing desk — no conflict of interest
• Partial fills possible (your order might fill across multiple providers at slightly different prices)

STP (Straight-Through Processing) is similar to ECN but with a distinction: the broker routes your order to its liquidity providers rather than into an open pool. You still get market prices without a dealing desk, but you see only the prices the broker's specific LPs provide, not a broader market.

Spread Mechanics in Detail

The spread isn't just a number — it's a dynamic reflection of market conditions, and understanding its mechanics can save you significant money over time.

Raw spread is what the interbank market actually quotes. On EUR/USD during London session, raw spreads from tier-1 banks can be as low as 0.0-0.1 pips. This is the "wholesale" price of liquidity.

Retail spread is what you see in your platform. It includes the raw spread plus your broker's markup. A broker might take a 0.1-pip raw spread and add 1.0 pip, showing you a 1.1-pip spread. Or they might show the raw spread and charge a $7 commission per standard lot round-turn. Compare total cost (spread + commission) to evaluate.

What affects spread width:

• Time of day: Spreads are tightest during London-New York overlap (highest liquidity). They widen during the Asian session, and blow out around the daily rollover (5 PM Eastern) when banks recalculate swap rates.
• News events: Before major releases (NFP, FOMC, ECB decisions), liquidity providers pull their quotes. Spreads can jump from 0.5 pips to 5-20 pips in seconds. This is liquidity risk, not broker manipulation.
• Market volatility: During flash crashes or unexpected events (SNB unpegging CHF in 2015, Brexit vote), spreads can widen to 50+ pips. Some brokers showed EUR/CHF spreads of 100+ pips during the SNB shock.
• Currency pair: Major pairs have the tightest spreads. Exotics like USD/TRY or USD/ZAR routinely carry spreads of 10-50 pips because liquidity is thinner.

Slippage: Why Your Fill Isn't Your Click

Slippage is the difference between the price you saw when you clicked and the price you actually got filled at. It's one of the least understood costs in trading.

Positive slippage (price improvement) happens when the market moves in your favor between your click and execution. You wanted to buy at 1.0850 and got filled at 1.0848. This is great — and reputable brokers pass this improvement to you.

Negative slippage happens when the market moves against you. You wanted to buy at 1.0850 and got filled at 1.0853. This is common during fast-moving markets.

Asymmetric slippage is a red flag. Some brokers pass negative slippage to traders but keep positive slippage for themselves. This is difficult to detect on individual trades but becomes visible over a large sample. If you notice that your fills are consistently worse than the price you clicked, request execution statistics from your broker. EU-regulated brokers are required to publish these.

Factors affecting slippage:

• Order type: Market orders experience more slippage than limit orders. Limit orders only fill at your price or better.
• Execution speed: The time between your click and the broker processing your order. ECN/STP execution is typically faster than dealing-desk execution.
• Market conditions: High-volatility periods produce more slippage because prices move faster than orders can be processed.
• Order size: Larger orders are harder to fill at a single price. A 10-lot order might fill partially at 1.0850 and partially at 1.0851.

Last Look and Execution Quality

Last look is a practice where liquidity providers reserve the right to accept or reject your order after seeing it, but before executing it. Originally designed to protect banks from latency arbitrage (ultra-fast traders exploiting stale quotes), last look gives LPs a window (typically 25-200 milliseconds) to decide if they want to fill your order.

The concern: during this window, if the price has moved against the LP, they reject. If it moved in their favor, they fill. This asymmetry means the LP always wins. The FX Global Code of Conduct has pushed for more transparency around last look, but it remains a common practice.

As a retail trader, you're generally shielded from last-look dynamics by your broker. But if you're trading through a prime-of-prime or directly accessing institutional liquidity, check whether your venues operate with or without last look.

Latency and Its Impact

In institutional trading, latency — the time it takes for data to travel between your system and the execution venue — matters enormously. High-frequency trading firms spend millions co-locating servers in the same data centres as exchanges to shave milliseconds off their execution times.

For retail traders, latency matters less but isn't irrelevant. If your broker's server is in London and you're trading from Sydney, every order has an inherent delay from the round-trip data travel. Some brokers offer regional servers or VPS (Virtual Private Server) hosting close to their servers to reduce this lag.

A good rule: if you're trading strategies that hold positions for hours or days, latency is irrelevant. If you're scalping on 1-minute charts, latency can be the difference between a winning and losing strategy. Choose your broker and connection infrastructure accordingly.

Practical Implications for Your Trading

• Know your broker's model. Ask whether they A-book or B-book. Reputable brokers will tell you. If they refuse to answer, that's your answer.
• Compare total costs, not just spreads. A broker advertising "0.0 pip spreads" with $7 commission per side ($14 round-turn) costs more per trade than a broker with 0.8-pip spreads and no commission on EUR/USD.
• Use limit orders for entries when possible. Limit orders eliminate negative slippage and can provide positive slippage (price improvement).
• Avoid trading during liquidity gaps. The daily rollover (5 PM ET), pre-news windows, and Sunday open are when execution quality is worst.
• Request execution reports. EU-regulated brokers (FCA, CySEC) must publish RTS 27/28 reports showing execution quality. Review them before committing to a broker.
• Test with small size first. Before scaling up, trade small positions and check your fill quality. Are your limit orders getting filled? Is slippage reasonable? Do you see requotes?

Understanding market microstructure won't make you a better technical analyst, but it will make you a more efficient trader. The edges you gain from better execution — tighter spreads, less slippage, avoiding liquidity traps — compound over thousands of trades into a meaningful difference in your bottom line.

Next lesson: the unglamorous but critical topic of taxes and legal compliance — because none of your profits matter if you don't handle them correctly.