Chapter 35 — Execution mechanics: spreads, slippage, financing

A strategy is a gross-return idea. Your account keeps the net. The gap between the two is execution cost, and on a leveraged 24-hour instrument it is larger and more variable than newcomers expect. This chapter is about the three leaks — spread, slippage, financing — and how to stop them quietly draining an edge that looks fine on paper.

The spread

The spread is the difference between the bid and the ask — the instant, guaranteed cost of a round trip. On XAUUSD it's usually quoted in cents (e.g., a 20-cent spread means $0.20 between buy and sell on the per-ounce price).

It is not constant. The spread is tightest during the London/New York overlap (deepest liquidity) and widens — sometimes dramatically — in three predictable windows:

The daily rollover (around 22:00–23:00 UTC, the broker's "end of day"), when liquidity thins.
Around high-impact news (FOMC, NFP, CPI): spreads can blow out 5–20× for seconds.
The weekend gap and Sunday re-open, when only thin liquidity is posted.

The lesson: a setup that's marginal on a 20-cent spread is a loser on a 2-dollar spread. Don't enter into a known spread-widening window expecting normal costs.

Slippage

Slippage is the difference between the price you expected and the price you got. It shows up two ways:

On entry/exit with market orders — your order fills at the next available price, which in a fast market is worse than the screen.
On stop-loss execution — and this is the dangerous one. A stop is a market order triggered at your level. In a fast move (a news spike, a liquidity sweep), it fills at the next price, which can be well beyond your stop. Your "1R risk" becomes 1.5R or 2R.

This is why the four-ruin-modes chapter (Ch 31) warns that a "5% risk" position can produce a 10% loss. The stop didn't fail — it slipped. The defenses:

Never hold a tight stop through a scheduled high-impact release. Either flatten before, or widen the stop and cut size so the slip-adjusted loss stays bounded.
Size for the worst fill, not the screen fill. Assume your stop slips in fast conditions; if that breaks your risk budget, the position is too big.
Prefer limit entries at predefined levels over chasing with market orders. Per the events chapter, the first 60 seconds after a print are algorithmic — don't click through it.

Financing (the swap) — the silent killer of slow trades

Covered in Chapter 34, but it earns repetition because it's the leak traders notice last. A leveraged CFD position is borrowed; you pay overnight financing on the full notional, every night, long or (often) short.

Worked example. You're long $200,000 notional of XAUUSD. At a financing rate of, say, 6% annualised on the notional, that's roughly:

$200,000 × 6% / 365 ≈ $33 per night

Hold three weeks (21 nights) and that's ~$690 — on a position where your entire planned profit might be 2R on a $1,000 risk, i.e., $2,000. The swap just ate a third of the trade before price moved. Triple the notional and the carry can exceed the edge entirely.

This is the mechanical reason Chapter 34 says: macro-horizon trades belong in futures or ETFs, not CFDs. The futures curve charges carry once per roll; the ETF charges a trivial annual fee; the CFD charges every single night on the whole notional.

Contract specs that bite

Know these before you size, not after:

Tick value. On GC, one tick ($0.10) = $10 per contract. On a CFD it depends on your lot definition. Mis-reading tick value is how people accidentally size 10× intended.
Minimum stop distance. Some brokers reject stops inside a minimum distance from price — relevant for tight scalps.
Margin and margin calls. Leverage cuts both ways; know the maintenance margin and the price move that triggers a call, especially across the weekend gap.
Weekend gap risk. Gold closes ~22:00 UTC Friday and re-opens ~22:00 UTC Sunday. A weekend headline (geopolitical, central-bank) can gap price past your stop with no chance to exit. This is why the discipline rule is: no oversized leveraged positions held naked over the weekend.

Figure 35.1 — Typical 24-hour XAUUSD spread profile

Figure 35.1 — Where the spread lives during the day. A 24-hour profile of typical XAUUSD spread, flat and tight through London/NY overlap, spiking at the daily rollover and around the NY-morning data window. The trade windows you want are the troughs; the windows that quietly tax you are the spikes.

On goldintel today

The brief's catalyst window and the economic-calendar panel exist precisely to keep you out of the spread/slippage spikes — when a high-impact event is inside 30–60 minutes, the brief flags it and sizing drops to "no new positions." Treat that flag as an execution-cost warning, not just a directional one: even a correct entry taken into the spike pays the widened spread and risks the slipped stop.

Common mistakes

Backtesting/journaling on the mid price and ignoring spread + slippage, so the live account underperforms the "strategy."
Holding leveraged CFDs for weeks and blaming the strategy when financing was the leak.
Tight stops through NFP/CPI/FOMC — the textbook way to turn a 1R risk into a 2R loss via slippage.
Mis-reading tick value and sizing an order an order-of-magnitude wrong.
Naked leverage over the weekend with a stop that can't protect against a Sunday gap.

Key takeaway

Your edge is measured after spread, slippage, and financing — trade the liquid windows, never hold a tight stop through scheduled news, and put slow trades in vehicles that don't charge nightly carry.