MetaTrader 5 broker fees: spread and commission data

This distinction changes the cost analysis. A trader comparing MT5 accounts is not comparing one software fee against another. The comparison is between pricing models attached to the same platform shell. The measurable variables are spread in pips, commission per standard lot, overnight financing, slippage distribution, rejection rate, and execution latency.

Platform software is not the fee source

MetaTrader 5 is a multi-asset platform. It supports forex, CFDs, futures, equities, and exchange-traded instruments where the broker enables them. The software does not set the economic terms of EUR/USD, GBP/JPY, XAU/USD, or an index CFD. Those terms sit in the broker’s symbol specification and account configuration.

For forex accounts, the visible cost usually falls into two structures:

The table is not a ranking. It is a decomposition. ECN does not dominate Market Maker pricing by definition. A raw spread plus commission can be cheaper, equal, or more expensive than a wider commission-free spread. The result depends on trade size, pair, trading frequency, fill quality, and holding period.

A standard lot in spot FX is 100,000 units of the base currency. On many USD-quoted major pairs, one pip on one standard lot is approximately $10. This convention gives a direct bridge between spread and commission.

A $7 commission per side is $14 round turn. On a one-lot EUR/USD trade, that is approximately 1.4 pips before spread. A $3 commission per side is $6 round turn, or approximately 0.6 pips before spread. Add the raw spread. The all-in cost is the only useful metric.

MT5 does not charge the trade. The broker’s account type charges the trade.

Market Maker spread-only accounts: cost is embedded in the quote

A Market Maker MT5 account often displays no separate commission line. That does not mean zero cost. The spread is the fee channel. The broker quotes a bid and ask. A long position enters at ask and exits at bid. The distance is paid immediately as negative mark-to-market.

For major pairs, spread-only accounts often show spreads in the 0.5–1.5 pip range. That range is not stable across all market states. It compresses during liquid sessions. It expands around rollovers, news releases, illiquid crosses, and abnormal volatility.

The spread-only model has one advantage for measurement: the transaction cost is visible at entry. It has one disadvantage: broker markup is not separated from market spread. The account statement may show no commission, but realized trading cost is still inside execution price.

For a system auditor, the cost equation is direct:

• Entry cost equals quoted spread at the time of fill.
• Exit cost is embedded in the closing quote.
• Total spread drag rises linearly with trade count.
• Strategy expectancy must exceed average spread plus slippage.
• Backtests using fixed low spreads understate drawdown and overstate profit factor.

The last point matters for algorithmic systems. A scalper with an average target of 3 pips cannot treat a 1.2-pip spread as a minor variable. The spread consumes 40% of gross target before slippage. If the strategy also has a stop of 3 pips, the break-even win rate moves sharply.

A simple example:

This is not a forecast. It is arithmetic. Symmetric gross targets and stops become asymmetric after spread. The spread shifts the distribution against the strategy. A backtest that ignores this will show a false edge.

Market Maker pricing can still be operationally usable. Lower-frequency systems may not care about a 1-pip difference if average trade duration is days and average target is 150 pips. The same spread can destroy a high-turnover intraday model. Cost relevance is a function of turnover, not account label.

ECN accounts: raw spread is not the final price

ECN-style MT5 accounts usually advertise raw spreads from 0.0 pips. That number is incomplete without commission. The typical commission range is $3–$7 per standard lot per side. The round-turn cost is twice the per-side number.

The all-in spread equivalent can be expressed as:

All-in cost in pips = raw spread in pips + round-turn commission in pip equivalent.

For one standard lot on a USD-quoted pair where one pip is approximately $10:

This is the core of any MT5 broker spread comparison. A raw spread of 0.0 pips with a high commission can be more expensive than a 0.8-pip spread-only quote. A raw spread of 0.2 pips with low commission can be cheaper than a 1.2-pip spread-only account.

The second issue is fill quality. ECN-style accounts may expose the trader to variable liquidity depth. A quoted 0.0-pip spread does not guarantee full-size execution at that price. Market orders can fill across levels. The cost then appears as slippage, not commission.

For automated systems, ECN account evaluation needs a dataset with these fields:

• Timestamp at order submission.
• Timestamp at broker acknowledgment.
• Timestamp at fill.
• Requested price.
• Filled price.
• Volume requested.
• Volume filled.
• Symbol spread at submission.
• Commission charged.
• Swap charged if held overnight.
• Order type and execution mode.
• Rejection or requote flag where applicable.

Without these fields, “best MT5 broker spreads” is a marketing phrase. It is not a measurable result. A usable comparison requires distribution analysis. Median spread is insufficient. The 95th percentile spread matters. Maximum adverse slippage matters. Commission variance by symbol matters. Time-of-day clustering matters.

A raw spread is a quote condition. An all-in cost is a realized condition.

Commission changes strategy geometry

Commission is linear by volume. It does not scale with signal quality. A system that trades ten standard lots pays ten times the commission of a one-lot system. The expected edge must scale accordingly.

A simple audit model can use expected value per trade:

EV = gross edge − spread − commission equivalent − average slippage − financing cost.

If gross edge is 1.4 pips and the all-in transaction cost is 1.1 pips, the residual is 0.3 pips before error. A small change in slippage standard deviation can erase the edge. If average adverse slippage is 0.2 pips and latency increases during volatility, the realized EV approaches zero.

This is why commission rates cannot be read in isolation. A $3-per-side account with unstable execution may be inferior to a $5-per-side account with lower slippage variance. The lower stated fee wins only if realized fills confirm it.

Spread comparison requires normalization

A direct forex broker MT5 fees comparison must normalize by pair, session, lot size, and account currency. Otherwise the sample is contaminated.

EUR/USD at London-New York overlap is not comparable with AUD/NZD near rollover. A one-lot trade is not comparable with a micro-lot trade if the broker uses minimum ticket charges. A USD account trading USD-quoted pairs is easier to audit than a non-USD account where conversion effects enter the statement.

A practical comparison protocol uses fixed parameters:

1. Same symbol set. Include at least EUR/USD, GBP/USD, USD/JPY, AUD/USD, USD/CAD, and one liquid cross. Do not mix majors with exotic pairs in a single average.

2. Same measurement windows. Record spreads during the Asian session, London open, London-New York overlap, and daily rollover. The average alone will hide tail costs.

3. Same account type. Compare raw-spread ECN accounts with raw-spread ECN accounts. Compare spread-only accounts with spread-only accounts. Mixed comparisons need all-in conversion.

4. Same order size. Use identical lots. Execution depth changes when size increases.

5. Same order type. Market orders, limits, and stops produce different slippage profiles. A broker can look efficient on limit orders and weak on stop execution.

6. Same holding period. Intraday cost analysis can ignore swap only if all positions close before rollover. Position systems cannot.

7. Same calculation basis. Convert commission into pip equivalent and add it to observed spread. Then compute mean, median, standard deviation, and tail percentiles.

The output should not be a single rank. It should be a cost distribution. For systematic traders, the difference between a mean all-in cost of 0.8 pips and 1.0 pips may be less important than the difference between a stable 99th percentile and a wide tail during high-impact data releases.

A cost comparison also needs a drawdown sensitivity test. Increase cost assumptions by 0.2, 0.5, and 1.0 pip. Re-run the system. If profit factor collapses under a 0.5-pip increase, the strategy is broker-sensitive. That is not a broker problem by itself. It is a fragility measurement.

Swap fees: overnight cost is a separate curve

Swap fees, also called rollover fees, apply to positions held overnight. They are based on the interest rate differential between the two currencies in the pair, then adjusted by broker policy. The charge or credit appears after the broker’s rollover time.

Swap is not part of spread. It is not part of commission. It is a financing term. For intraday systems it may be zero if no position crosses rollover. For swing systems and carry-sensitive models it can dominate trade cost over time.

The swap impact is path-dependent. A position held for one night has one financing event. A position held for twenty nights has twenty events, subject to broker schedule and triple-swap conventions where applicable. The longer the holding period, the less useful a spread-only comparison becomes.

A long-term MT5 position audit separates four components:

A position strategy can tolerate higher entry spread if swap terms are favorable. The reverse is also true. Low spreads cannot repair persistent negative financing on a position held through multiple rollover events.

This is relevant for pairs with large rate differentials. It is also relevant when brokers apply different long and short swap rates on the same pair. The trader’s direction matters. A long signal and a short signal do not have symmetric financing unless the broker’s terms make them symmetric, which is uncommon.

The audit should therefore record realized swap, not just published swap. Published contract specifications can change. Account history is the final dataset.

Execution type can convert visible fees into hidden costs

MT5 supports several execution types, including Instant, Request, Market, and Exchange execution. The broker chooses what applies to symbols and account types. Execution type affects how orders are accepted, filled, rejected, or requoted.

The nominal fee schedule can look identical across two brokers. Realized cost can diverge through execution mechanics.

Market execution usually fills at available market prices. The fill can be better or worse than the requested level. Slippage is part of the cost distribution. Instant or Request execution may introduce requotes or rejection behavior when the requested price is no longer available. Exchange execution applies where the broker routes orders to an exchange venue for listed instruments.

For FX traders, the practical variables are:

• Latency. Time from order submission to fill. Higher latency increases exposure to price movement.
• Slippage mean. Average difference between requested and filled price.
• Slippage standard deviation. Variability of the fill error.
• Rejection rate. Share of orders not accepted under stated conditions.
• Partial fill behavior. Relevant for larger tickets and thin liquidity.
• Stop execution gap. Difference between stop trigger and final fill.

A low metatrader 5 commission rate does not compensate for poor execution if slippage variance is high. The reverse can also occur. A higher commission account can produce lower total cost if fills are stable and the spread tail is narrow.

Execution cost becomes visible in stop-heavy strategies. Stop orders are not insurance against exact price. They are instructions that convert into execution when conditions are met, subject to broker rules and market state. During volatility, the fill may deviate from the stop level. The loss distribution must include this.

For systems using expert advisors, latency is not cosmetic. A 200-millisecond difference may be irrelevant for a four-hour breakout system. It may be material for a news straddle or a tick-driven mean-reversion model. The broker test must match the strategy’s time scale.

The usable fee metric is all-in realized cost

A clean MetaTrader 5 broker fee comparison reduces every pricing format to one unit: realized cost per trade, expressed in pips or account currency.

The basic calculation for a closed trade is:

• Spread cost: bid-ask cost at entry and exit, or effective price difference versus mid where measurable.
• Commission: total entry plus exit commission.
• Slippage: filled price versus expected execution price.
• Swap: overnight debit or credit.
• Conversion: account-currency adjustment where relevant.

Then aggregate by symbol and strategy type. A broker can be efficient on EUR/USD and expensive on GBP/JPY. A broker can be efficient for intraday trades and inefficient for positions held over rollover. A broker can be acceptable for manual swing execution and unsuitable for latency-sensitive automation.

The correct comparison unit is not “lowest spread.” It is:

All-in cost per round turn, by symbol, at the actual traded size, during the actual traded session.

That number can then be mapped to strategy expectancy. If a system’s average gross profit per trade is 8 pips and all-in cost is 0.8 pips, cost consumes 10% of gross profit. If gross profit is 2 pips and all-in cost is 0.8 pips, cost consumes 40%. Same broker. Different viability.

A robust broker review should also report dispersion:

This is the difference between a fee table and a trading-cost model. Fee tables are static. Trading costs are distributions.

Risk-reward summary and backtest limits

MT5 broker fees are broker-defined. MetaTrader 5 is not the charging entity. The relevant split is spread-only pricing versus raw spread plus commission, with swap and execution effects added after that.

For Market Maker-style accounts, major-pair spreads commonly fall around 0.5–1.5 pips. The cost is embedded in the quote. For ECN-style accounts, raw spreads may start near 0.0–0.3 pips, while commission commonly ranges from $3 to $7 per standard lot per side. The round-turn pip equivalent can erase the apparent advantage of the raw quote.

The strict comparison is mechanical:

• Convert commission into pips.
• Add observed spread.
• Add average slippage.
• Add swap for overnight trades.
• Test cost sensitivity against strategy expectancy.
• Reject any backtest that assumes fixed low spread without slippage tails.

A broker with lower stated costs is not automatically superior. A broker with higher stated commission is not automatically inferior. The decision depends on realized all-in cost, execution variance, and the strategy’s margin of safety.

Backtest limits remain binding. Historical spread files may not match live broker spreads. Commission schedules can change. Liquidity provider markups are not fully disclosed. Swap rates can move. Latency in a test environment may differ from live routing. Therefore the final broker selection should be based on live or forward-test execution logs, not on published spread claims alone.