Choosing a Broker for Scalping: My Trial and Error Story

Scalping is unforgiving about broker quality in a way that other styles aren't. If you're a swing trader holding positions for days, a 0.5-pip difference in spread is noise. If you're a scalper targeting 5-8 pips per trade, that same 0.5-pip difference is 6-10% of your target profit. Every single trade. It compounds fast.

I didn't fully understand this when I started. I chose my first broker the way most beginners do — based on a review site and a well-designed website — and paid the price over the following six months in costs I didn't fully account for.

Broker One: The One That Said It Allowed Scalping (But Didn't Really)

My first broker — I won't name them — advertised "scalping allowed" prominently on their website. What this turned out to mean in practice was: scalping is technically permitted, but we'll make it unprofitable through execution.

Specifically: requotes. On every second or third fast entry, I'd get a requote — the platform would tell me the price had moved and ask if I wanted to fill at the new price. By the time I confirmed, the price had often moved again. For a scalper trying to enter a position within a specific 1-2 pip window, this made the strategy unworkable.

I contacted support. They explained that during high-volatility periods, requotes were "unavoidable." The thing is, USD/JPY during the Tokyo session open isn't especially high-volatility compared to, say, a London-New York overlap during NFP. The real issue was execution model — this broker was clearly a market maker routing my orders in a way that prioritised their interests over fast fills.

After three months and a modest loss, I moved on.

Broker Two: Great Spreads, Wrong Session

The second broker had genuinely excellent spreads — I verified this myself by running a demo account and logging every spread on USD/JPY at five-minute intervals for two weeks. During European and US hours, they were excellent. During the Asian session, when I actually trade, they were 40-50% wider.

This is a legitimate thing to check and one I'd never thought about before. The "typical spread" figures brokers quote are usually from their best conditions — often during peak London or New York liquidity. If you trade the Asian session, or overnight from a US perspective, you're in thinner market conditions, and some brokers reflect that with wider spreads while others maintain tighter spreads through better LP relationships.

The broker was otherwise fine. Just not optimised for my hours. I still recommend them to friends who trade during London session.

Broker Three: The Hidden Cost I Missed

Third broker was a popular name — good reputation, FCA regulated, genuine ECN execution. Raw spreads were excellent. Commission was listed as $3 per lot per side.

What I didn't calculate properly until two months in: at my trading volume (sometimes 15-20 trades per day on micro and mini lots), the commission was adding up to a significant daily cost. When I modelled it properly against my average trade size and win rate, the math only worked if I was hitting my targets consistently, leaving no room for typical variance in a strategy's performance.

The broker wasn't doing anything wrong. I'd just failed to model my actual all-in cost before committing. This is basic, and I got it wrong. The spreadsheet I built after this mistake — tracking every trade's full cost including commission, swap, and spread — is something I still use today.

Broker Four: The Right Fit (Finally)

I ended up with IC Markets on their cTrader raw account, and I've been there for over a year now. Here's why it works for me specifically as a scalper on the Asian session:

Execution is fast and requote-free. cTrader's execution engine is well-regarded for low latency. I've had perhaps two requotes in twelve months of active scalping. It's genuinely different from my first broker's experience.

Asian session spreads are acceptable. Not the 0.0-0.1 you see during peak London hours, but typically 0.3-0.5 on USD/JPY during the Tokyo open — workable for my target sizes.

Commission is fair at my volume. $3 per lot per side. Because I've now built my cost model properly and size my trades accordingly, this works with my strategy's expectancy.

No restrictions on scalping. IC Markets explicitly allows scalping and EAs. There's no vague "at our discretion" language in the terms. The platform handles fast exits and entries without pushback.

Negative balance protection. Scalping with leverage on fast-moving currency pairs carries real risk of a flash crash eating through your margin faster than you can react. Having negative balance protection capped my worst-case loss at my account balance rather than something greater.

What I Now Check Before Considering Any Broker for Scalping

• Is scalping explicitly permitted? Not implied, not technically not prohibited. Explicitly stated as allowed in the terms.
• What is the execution model? ECN/STP preferred. Market maker execution is incompatible with serious scalping.
• What are the spreads during MY trading hours? Not peak hours. The actual hours I'll be in the market.
• What is the commission, and does my strategy's expectancy support it? Model this before opening an account, not after.
• Does the platform support fast order modification? Scalping often requires rapid adjustments to stops and targets. The platform has to handle this without lag.
• What is the server latency from my location? If possible, check the broker's server location and test ping from your trading setup. Milliseconds matter for scalping.

My full setup is now: IC Markets raw cTrader account, VPS hosted in a Singapore data centre (close to IC Markets' servers), and a fixed strategy with pre-calculated entry and exit levels before each session. The broker is no longer a variable in my performance — it's a constant I've optimised and moved past.

That's the goal. Your broker should be boring. If you're thinking about your broker while you're in a trade, something is wrong.