Methodology
How we test, score, and rank CFD brokers—transparently and reproducibly.
1. Spread Measurement
We record spreads at three fixed times each trading day: London open (08:00 GMT), New York open (13:30 GMT), and Asian session mid-point (02:00 GMT). Measurements are taken from live accounts, not demo. Each broker is measured across 5 instruments: EUR/USD, US100, XAU/USD, US Oil, and BTC/USD.
2. Execution Quality
We place 50 market orders per broker per month and record: fill time (ms), slippage vs quoted price, and rejection rate. Orders are placed during high-volatility windows (NFP releases, API inventory) and normal market conditions.
3. Cost Scoring (30%)
Derived from average spread + overnight swap + inactivity fees. Lower is better. We normalise across asset classes so a forex specialist is not penalised for wider crypto spreads.
4. Execution Scoring (25%)
Weighted combination of fill speed, slippage, and rejection rate. DMA/ECN brokers receive a bonus for transparent order routing. Market makers are not penalised if slippage is within advertised bounds.
5. Trust & Regulatory (20%)
Points for: FCA/ASIC/CySEC regulation (+3 each), negative balance protection (+2), segregated client funds (+2), verified execution model (+2). Penalties for: recent regulatory fines (−2 to −5), offshore-only structure (−3).
6. Asset Range & UX (15%)
Count of tradeable CFD markets, platform stability (uptime %), mobile app quality, and charting tools. We do not score bonuses or gamification features positively.
7. Quarterly Re-Test
Every score is stale-dated. If a broker has not been re-tested within 90 days, its score is marked with a warning icon and excluded from "Best for" recommendations until refreshed.