MQL5 Spread Filter: How to Control Spread Before Order Execution

1. What Is an MQL5 Spread Filter?

Bottom line:
A spread filter is a condition that allows an order only when the current spread is at or below a set value. It is one of the most basic ways to control trading costs.

Definition:
The spread is the difference between the Bid price and the Ask price. It is a real cost that exists from the moment a trade is opened.

1.1 Definition of a Spread Filter

A spread filter is a conditional branch in an EA (Expert Advisor) that prevents entries under unfavorable market conditions.

In practical terms, the logic works like this:

• Small spread -> allow entry
• Wide spread -> block entry

This mechanism helps you avoid unnecessary costs and stabilize the expected value, or average profit, of your trades.

1.2 Why a Spread Filter Is Needed

The spread is not just a number. It is the initial loss of the trade itself.

For example:

• 2.0-pip spread -> -2.0 pips immediately after entry
• 0.5-pip spread -> -0.5 pips immediately after entry

In other words, if you enter while the spread is wide,
the same strategy can suffer a much worse win rate and profit margin.

Spreads often widen sharply in these situations:

• Economic news releases
• Right after the market opens, such as at the start of the week
• Low-liquidity periods, such as early morning or late night

For this reason, a spread filter becomes
an important control mechanism that automates the decision not to trade.

1.3 Related Terms for Beginners

To make the concept easier to understand, here are the related terms in simple form.

• spread
  -> The difference between the buy price and sell price. It is a trading cost.
• slippage
  -> The difference between the requested order price and the actual execution price.
• execution
  -> The process in which an order is filled in the market.
• order condition
  -> A condition used to execute an entry, including filters.

1.4 Where a Spread Filter Fits in EA Design

Inside an EA, a spread filter is placed in this part of the flow:

Signal generation -> Filter (spread, etc.) -> Order execution

This filter layer is very important. Common examples include:

• Spread filter
• Time filter, such as trading hour restrictions
• Volatility filter, such as ATR

In short, a spread filter is
part of the preprocessing that affects the quality of an EA.

1.5 Common Misunderstandings and Mistakes

Here are the points where beginners often get stuck.

• Assuming the spread can be ignored
  -> In live trading, it is one of the most important cost factors.
• Assuming it is fine because the backtest looks good
  -> Many testers use a fixed spread, which can differ from live conditions.
• Thinking a filter always reduces profit
  -> In practice, removing unnecessary trades can improve expected value.

1.6 Relationship With Other Methods

A spread filter is a form of pre-entry cost control.

• Spread filter -> controls cost before entry
• Slippage control -> handles price deviation during execution

Because these two have different roles, using them together is the standard approach.

2. How to Implement a Spread Filter in MQL5

Bottom line:
In MQL5, you can implement a spread filter by getting the current spread, comparing it with a threshold, and controlling execution before sending the order.

Definition:
Getting the spread means retrieving the current difference between Ask and Bid as a numeric value and using it in a conditional branch.

2.1 Overall Implementation Steps

You can implement it in four steps:

• 1. Get the current Bid / Ask
• 2. Calculate the spread
• 3. Compare it with the allowed spread threshold
• 4. Execute the order only if the condition is met

This structure is common to all EAs.

2.2 Basic Code for Getting the Spread

double bid = SymbolInfoDouble(_Symbol, SYMBOL_BID);
double ask = SymbolInfoDouble(_Symbol, SYMBOL_ASK);

// Spread in points
double spread = (ask - bid) / _Point;

Explanation: Key Points

• _Symbol -> The current currency pair
• _Point -> The minimum price unit, such as 0.00001
• (ask - bid) -> The spread itself

Many brokers handle this in points, so
it is more practical to manage the filter by points rather than converting everything to pips.

2.3 Implementing the Filter Condition

double maxSpread = 20; // Allowed spread in points

if(spread > maxSpread)
{
    // Spread is too wide -> do not enter
    return;
}

Key Points

• Use return; to stop the process and prevent unnecessary orders.
• The threshold must be optimized for each strategy.

2.4 Practical EA Integration Example

void OnTick()
{
    double bid = SymbolInfoDouble(_Symbol, SYMBOL_BID);
    double ask = SymbolInfoDouble(_Symbol, SYMBOL_ASK);
    double spread = (ask - bid) / _Point;

    double maxSpread = 20;

    // Spread check
    if(spread > maxSpread)
        return;

    // Check the signal here, for example
    if(CheckBuySignal())
    {
        // Order processing
        trade.Buy(0.1);
    }
}

Practical Points

• Checking immediately before the order is important
  -> The spread can change after the signal appears.
• Evaluate it on every OnTick() call to reflect the latest market state.

2.5 Common Mistakes and Fixes

1. Mistakes With Spread Units

• Bad: confusing pips and points
• Fix:
  -> Calculate consistently with (_Point).

2. Using a Threshold That Is Too Fixed

• Bad: always using the same value, such as 20
• Problem: the best value differs by currency pair and time of day.

Fix:

• Set values by currency pair.
• Switch values by time period.

3. Gap Between Backtesting and Live Trading

• Bad: results look good in the tester but break in live trading.
• Cause: fixed spread assumptions.

Fix:

• Test with variable spreads.
• Validate with live account logs.

2.6 Why This Implementation Works

Because the spread is an entry cost,
the structure is as follows:

Net profit/loss = Strategy profit - Spread - Slippage

In other words:

• Wide spread
  -> Higher initial loss
  -> Worse breakeven point

Therefore,
removing these entries before execution can improve expected value.

2.7 Differences From Other Methods

• Spread filter
  -> Cost control before entry
• Slippage control
  -> Control of price deviation during execution
• Limit order
  -> Attempts execution at a favorable price

Because each one has a different role, using them together is the standard approach.

3. How a Spread Filter Works and Why It Helps

Bottom line:
A spread filter improves expected value by controlling the initial cost. It optimizes the profit and loss structure itself, not simply the win rate.

Definition:
Expected value is the average profit per trade. It is determined by
(win rate x average profit) – (loss rate x average loss) – costs.

3.1 How the Spread Affects Trading

The spread is a hidden loss, but in practice it is a cost that is already locked in from the start.

Example:

• 1.0-pip spread -> starts at -1.0 pips
• 3.0-pip spread -> starts at -3.0 pips

This difference may look simple, but in live trading it has these effects:

• The distance to take profit (TP) becomes longer.
• The distance to stop loss (SL) becomes shorter.
• The impact is larger for short-term strategies.

This can be especially severe for scalping.

3.2 Understanding It Through Expected Value

The spread directly affects the expected value formula.

Expected value = Win rate x Average profit - Loss rate x Average loss - Spread

The important points are:

• The spread is always subtracted.
• It is one of the few elements you can control.

In other words:

It is a lever you can improve without changing the strategy itself.

3.3 How the Filter Improves Expected Value

When you add a spread filter, the following change occurs:

Before, Without the Filter

• Entries are taken on every signal.
• High-spread periods are included.
• Unnecessary losses increase.

After, With the Filter

• Entries are taken only under better conditions.
• High-cost environments are excluded.
• The number of trades decreases, but trade quality improves.

Result:

• Improved PF, or profit factor
• Reduced DD, or drawdown
• Better reproducibility

3.4 Why Profit Can Improve Even When Trade Count Drops

This is a common beginner misunderstanding.

Intuition:

• Fewer trades -> less profit

Reality:

• Removing unfavorable trades
  -> higher average profit

In short:

Profit = Frequency x Quality

A spread filter is a method for improving quality.

3.5 Why Spreads Widen: Market Structure

Spreads are not random. They change based on market structure.

Main causes:

• Lower liquidity, meaning fewer orders
• Higher volatility, meaning sharper price movement
• Dealer risk management, meaning broker-side adjustment

Specific examples:

• During news -> spreads widen sharply
• Early morning -> lower liquidity widens spreads

For this reason,
the spread can be used as a risk signal.

3.6 Difference From Slippage

These two are often confused, so separate them clearly.

Bottom line:
Spread is prevention. Slippage is response.

3.7 Practical Use Cases

A spread filter is especially useful for:

• Scalping, or short-term trading
• Avoiding trades around economic releases
• Stabilizing an EA

On the other hand:

• Swing trading, or longer-term trading
  -> the impact is relatively smaller.

Note that the importance changes depending on the strategy.

3.8 Common Misunderstandings

• “The spread depends on the broker, so I cannot control it.”
  -> That is exactly why you need to control when you trade.
• “The spread is constant.”
  -> In practice, it changes. This is called a variable spread.
• “It reduces profit.”
  -> The main purpose is to reduce unnecessary losses.

4. Spread Filter Compared With Other Methods

Bottom line:
A spread filter focuses on removing entry cost before the trade. It has a different role from slippage control and time filters. Using them together is the expected design.

Definition:
The compared methods are groups of conditions, or filters, that control trade accuracy, safety, and cost.

4.1 Comparison of Major Filter Methods

The table below compares representative control methods used in practice.

4.2 Spread vs. Slippage

These are the two most commonly confused concepts.

Spread Filter

• Fixed cost that always occurs
• Can be judged before entry
• Relatively easy to avoid

Slippage Control

• Uncertain cost that depends on conditions
• Occurs during execution
• Difficult to avoid completely

Practical conclusion:

• Spread -> remove it in advance
• Slippage -> control it within an allowed range

4.3 Spread vs. Time Filter

Characteristics of a Time Filter

• Excludes specific times, such as early morning or economic releases
• Simple and lightweight

Problems

• It does not check the actual spread condition.
• There may be cases where spreads are wide even during normally safe hours.

Advantage of a spread filter:

• It is based on the actual market state and execution environment.
• It allows more accurate judgment.

In practice:

Time filter + Spread filter

is the standard combination.

4.4 Spread vs. Volatility Filter

Volatility, Such as ATR

• Measures the size of price movement.
• Useful for judging trend or range conditions.

Difference

• Spread -> cost indicator
• Volatility -> market state indicator

In other words:

• Spread -> “Should we trade?”
• Volatility -> “Which strategy should we use?”

4.5 Spread vs. Limit Order

Limit Order

• Executes only at a specified price.
• Can effectively avoid paying an unfavorable spread.

Problems

• Risk of no execution, which creates opportunity loss
• Risk of missing a trend move

Advantage of a spread filter:

• It keeps execution opportunities available.
• It controls only the condition.

4.6 Recommended Practical Configuration

Design it as a combination rather than relying on a single method.

Basic Structure

Signal
|
Time filter
|
Spread filter
|
Slippage control
|
Order execution

Reasons

• Layered risk control
• Reduced dependence on one condition
• Better reproducibility

4.7 Common Design Mistakes

• Trying to solve everything with only a spread filter
• Ignoring slippage
• Using filters in the wrong order

Especially important:

A design that does not check the spread immediately before the order is risky.

4.8 Which Filter Should You Prioritize?

Here is a strategy-based view:

• Scalping -> spread is the top priority.
• Day trading -> spread plus time filter.
• Swing trading -> time and volatility are more important.

Key decision factors:

• Trading frequency
• Profit target size
• Required execution accuracy

5. Best Spread Filter Values and Practical Use Cases

Bottom line:
The best spread value is not fixed. It must be adjusted by currency pair, time of day, and strategy. A single universal setting is inefficient.

Definition:
The best value means the spread threshold that maximizes expected value, or profit minus cost.

5.1 Recommended Spread Guidelines

General guidelines in points:

Note

• Values can differ greatly by broker and account type.
• ECN accounts are usually lower, while standard accounts are usually higher.

Bottom line:
Always collect logs in your own trading environment.

5.2 Best Settings by Strategy

Scalping, Very Short-Term Trading

• Recommended: strict, around 10-15
• Reason: cost has the largest impact.

The filter is essential.

Day Trading

• Recommended: moderate, around 15-30
• Reason: the profit target is larger, so some spread can be tolerated.

Swing Trading

• Recommended: looser, 30 or higher
• Reason: the spread has a relatively smaller impact.

5.3 Optimization by Time of Day

Spreads can change greatly depending on the time of day.

Risky Periods

• Early morning, when liquidity is low
• During economic releases
• At the start of the week

More Stable Periods

• London session
• New York session

Practical handling:

if(isHighRiskTime())
{
    maxSpread = 10;
}
else
{
    maxSpread = 20;
}

Key point:
Change the threshold dynamically by time period.

5.4 Adjustment by Currency Pair

Each currency pair has different characteristics.

Examples:

• EURUSD -> stable with a low spread
• GBPJPY -> higher volatility with a wider spread

Implementation example:

double maxSpread;

if(_Symbol == "EURUSD")
    maxSpread = 15;
else if(_Symbol == "GBPJPY")
    maxSpread = 30;

In practice, using an array or configuration file is recommended.

5.5 Dynamic Spread Filter, Advanced

This is a more advanced method.

Method

• Get the historical average spread.
• Compare it with the current value.

if(currentSpread > averageSpread * 1.5)
    return;

Advantages

• Adapts to market conditions.
• More accurate than a fixed value.

Disadvantages

• More complex implementation.
• Higher calculation cost.

5.6 Practical Use Cases

A spread filter is especially effective for:

• Stabilizing an EA, which is the most important use
• Scalping strategies
• Avoiding trades around economic releases

On the other hand:

• Long-term investing -> the effect is limited.

5.7 Common Mistakes and Improvements

1. Setting It Too Strictly

• Problem: fewer entry opportunities
• Fix: validate the best value with logs.

2. Setting It Too Loosely

• Problem: more unnecessary losses
• Fix: evaluate by PF and DD.

3. Trusting a Fixed Value Too Much

• Problem: it cannot adapt to market changes.
• Fix: use dynamic adjustment.

5.8 Basic Optimization Policy

Choose the best value based on:

• Maximizing PF, or profit factor
• Minimizing DD, or drawdown
• Balancing the number of trades

Do not optimize only for raw profit.
Evaluate by risk-adjusted return.

5.9 Practical Summary

• The spread is not fixed.
• The best value differs by strategy.
• Dynamic adjustment is the strongest approach.

6. Common Spread Filter Mistakes and Notes

Bottom line:
Adding a spread filter is not enough by itself. If you get the units, timing, or setting method wrong, it can work against you.

Definition:
A failure means a state where the filter is implemented but expected value does not improve, or becomes worse.

6.1 Unit Mistakes: Points vs. Pips

Problem

In MQL5, spreads are usually handled in points,
but beginners often confuse points with pips.

Example:

• 1 pip = 10 points for a 5-digit currency quote

Common mistake:

double maxSpread = 2; // Intended as pips, but actually 2 points

This means the filter will trigger almost all the time.

Fix

double maxSpread = 20; // Equivalent to about 2.0 pips for a 5-digit quote

Bottom line:
In MQL5, keep everything based on points.

6.2 Wrong Check Timing

Problem

This happens when the spread is checked only at signal generation.

Bad example:

if(CheckSignal())
{
    // Check only once here
}

Why It Is a Problem

• The spread can change after the signal occurs.
• The condition may no longer be valid at order execution.

Correct Design

// Check immediately before the order
if(spread > maxSpread)
    return;

Bottom line:
Always evaluate it immediately before the order.

6.3 Trusting a Fixed Value Too Much

Problem

Using the same spread value all the time:

double maxSpread = 20;

Why It Is a Problem

• Each currency pair has different characteristics.
• Spreads change greatly by time of day.
• Market conditions change.

Fix

• Set values by currency pair.
• Switch values by time period.
• Introduce a dynamic filter.

Bottom line:
A fixed value is fine for simple implementation, but it is often not enough for live trading.

6.4 Overreliance on Backtesting

Problem

This happens when decisions are made only from tester results.

Background

• Many testers use fixed spreads.
• Live trading uses variable spreads.

Impact

• Expected value can collapse in live trading.
• DD, or drawdown, can increase.

Fix

• Use variable-spread data.
• Validate with live account logs.

Bottom line:
Backtesting alone is not enough.

6.5 Judging Only by Spread

Problem

Entry decisions are made only from the spread.

What Gets Missed

• Slippage, or execution deviation
• Liquidity
• Volatility

Fix

Use a combined design:

Spread + Slippage + Time filter

Bottom line:
A single filter is fragile.

6.6 Filter Is Too Strict

Problem

• Entry opportunities drop sharply.
• Expected value can decrease.

Typical Example

• maxSpread = 5, which is overly strict

Fix

• Check the number of trades.
• Evaluate the balance with PF.

Bottom line:
The balance with opportunity loss is important.

6.7 Filter Is Too Loose

Problem

• Unnecessary trades increase.
• DD becomes worse.

Fix

• Analyze logs.
• Compare performance during high-spread periods.

Bottom line:
If the filter has no effect, the threshold may be too loose.

6.8 Not Taking Logs

Problem

• You cannot optimize the threshold.
• You cannot analyze the cause of poor results.

Fix

Print("Spread:", spread);

Recommended log items:

• Spread
• Whether entry was allowed
• Actual profit and loss

6.9 Practical Checklist

• Use points consistently.
• Check immediately before the order.
• Do not depend on a fixed value.
• Do not overtrust backtests.
• Use it together with other filters.

7. Frequently Asked Questions

Bottom line:
A spread filter is a simple conditional branch, but in live trading, its settings, units, and combined design can strongly affect results. This section answers the common questions.

Definition:
This FAQ summarizes concrete questions and answers that beginner to intermediate users often face during implementation and operation.

7.1 Is a Spread Filter Required?

A. It is not strictly required, but it is recommended for almost every EA.
In scalping especially, the spread directly reduces profit, so an EA without this filter may suffer a much worse expected value.

7.2 Should I Use Pips or Points for the Spread Unit?

A. In MQL5, using points consistently is the practical choice.
Because MQL5 works internally with points, using points helps prevent conversion mistakes.

7.3 What Is the Best Spread Value?

A. There is no fixed correct answer. It changes by currency pair, time of day, and strategy.
As rough guidelines:

• EURUSD: 10-20 points
• GBPJPY: 20-40 points

However, always collect logs in your own environment and optimize the value.

7.4 Is Backtesting Alone Enough?

A. No, it is not enough.
Many testers use fixed spreads, so they cannot fully reproduce the variable spreads of live trading.
Always confirm results with forward testing or live account logs.

7.5 Does a Wide Spread Always Cause a Loss?

A. Not always, but it puts the trade at a disadvantage.
The spread is an initial cost, so the wider it is, the farther the breakeven point becomes.
This is especially damaging for short-term trading.

7.6 Is a Spread Filter Enough by Itself?

A. No. Combining filters is the standard approach.
Recommended structure:

• Spread filter
• Slippage control
• Time filter

This can greatly improve stability.

7.7 Will a Stricter Filter Increase Profit?

A. Not always.
If the filter is too strict, entry opportunities decrease and profit may fall instead.
You need to judge the balance among PF, DD, and the number of trades.

7.8 Where Can I Check the Spread?

A. In MQL5, you can get it with the following code.

double spread = (SymbolInfoDouble(_Symbol, SYMBOL_ASK) 
               - SymbolInfoDouble(_Symbol, SYMBOL_BID)) / _Point;

You can also check it in MT5 Market Watch.

7.9 How Can I Confirm the Effect of a Spread Filter?

A. The most reliable method is to compare results with the filter on and off.

• Backtest with the filter ON and OFF.
• Compare results in forward testing.
• Analyze profit and loss during high-spread periods.

Bottom line:
Make the decision based on data.