User Guide : Quality Filter User Guide

Quality Filter User Guide

Version: 1.15 | Author: Ich Khiem Nguyen | MQL5: https://www.mql5.com/en/users/khiemni/seller

1. What is Quality Filter?

Quality Filter is a feature that helps you receive only the best signals — price patterns formed clearly, with balanced proportions and high reliability — and reject weak signals, dull patterns, and those below standard.

Think of yourself as a pageant judge: you don't eliminate contestants for being unintelligent or having a bad personality — you simply score the form of the price pattern (ratios, balance, market context) and select only patterns that meet minimum threshold.

Core Principles

• Every detected signal receives a quality score from  0.0  to  1.0
• 0.0  = very weak pattern, not clear
• 1.0  = perfect pattern, textbook beautiful
• You set a minimum threshold — only signals reaching or exceeding this threshold appear on the chart

Real-World Results

• Enable Quality Filter + high threshold → fewer signals, higher quality
• Disable Quality Filter → all signals appear regardless of score

---

2. Input Settings Guide

Quality Filter has 2 parameters located in the  === SIGNAL QUALITY FILTERS ===  group in the indicator's Input dialog.

Parameter 1: Enable/Disable Quality Filter

MT5 Input Name	Type	Default
Quality Score Filter (pattern strength)	bool (true/false)	Indicator-specific

• true  = Enable the filter — only signals reaching the threshold display
• false  = Disable the filter — all signals display regardless

Default values by indicator:

• Mirage (Fakey):  false  — disabled, receives more signals by default
• PinBar:  true  — enabled, default threshold 0.55 (fairly tight)
• Inverted Hammer:  true  — enabled, default threshold 0.30 (fairly loose)

Parameter 2: Minimum Quality Threshold

MT5 Input Name	Type	Default	Notes
Min Quality (0.0-1.0, 0.35=balanced)	double	0.30 – 0.55	Indicator-specific

• Enter value from  0.0  to  1.0
• Only takes effect when Quality Filter is ENABLED
• Recommended starting point:  0.35  (balance between quantity and quality)

---

3. How is Quality Score Calculated?

Each indicator has its own pattern detection engine. The quality score reflects how "perfect" the pattern is based on the following factors:

Candle Proportions

The engine measures candle components: body, upper wick, lower wick. A "beautiful" pattern per theory has specific ratios between components. Candles matching the theoretical ratio more closely score higher.

Market Context

Where a pattern appears on the chart also matters. The engine considers:

• Relative position compared to recent highs/lows
• Strength of the current trend
• Volatility (ATR) at the moment the pattern forms

Additional Factors

Depending on the indicator, the engine may also calculate:

• Potential Risk/Reward ratio
• Pattern clarity versus surrounding candles
• Confirmation candle strength (if applicable)

Important Note

Quality score does NOT predict price direction. A signal with Q = 0.90 doesn't mean price will definitely move in the right direction — it only means the pattern formed clearly and is technically beautiful.

---

4. Choosing the Right Threshold

Reference Table: Filter Ratio (H1 XAUUSD)

Threshold (Min Quality)	Signals Retained	Level
0.25	~75%	Very loose — almost no filtering
0.35	~47%	Balanced — good starting point
0.45	~19%	Tight — good signals only
0.55	~10%	Very tight — excellent signals only

Note: Ratios measured on H1 XAUUSD data. Different timeframes and pairs may show different ratios.

Threshold Selection by Purpose

You're just starting out, want to see many signals:

• Set threshold to  0.30  or disable Quality Filter entirely
• Purpose: observe and learn quality assessment

You want balance between quantity and quality:

• Set threshold to  0.35  (recommended default)
• Suitable for most traders

You want selective trading, fewer but more reliable setups:

• Set threshold to  0.45  –  0.55
• Suitable when reducing order frequency and increasing R:R

You scalp multiple times daily:

• Don't set threshold too high (>0.50), will lack signals
• Recommended:  0.30  –  0.40

Don't Set Threshold Too High

Threshold 0.70+ will eliminate nearly all signals under normal market conditions. Markets rarely create "absolutely perfect" patterns.

---

5. Reading Quality Score on Chart

After a signal appears, you can view the quality score in 2 places:

5a. Directly on Chart (at Signal Arrow)

If you enable  Show Quality Score = true  in the indicator's Visualization group, each signal displays a quality label right next to the arrow:

• Format:  Q: 0.72  or  72%
• Position: usually above or to the right of the arrow
• Color: usually matches arrow color (green = Buy, red = Sell)

5b. On Dashboard

The indicator's Dashboard panel lists active signals. The quality column displays the Q score of each signal for quick comparison across active trades.

Dashboard Not Showing Q Score?

• Check  Show Quality Score  in the Visualization group — may be disabled
• Some older indicator versions don't support direct Q display on chart

---

6. Frequently Asked Questions (FAQ)

Q1: Does disabling Quality Filter slow the indicator? No. Quality Filter is just a final check — pattern detection happens fully inside. Enabling/disabling has no significant performance impact.

---

Q2: Do signals filtered by Quality Filter display anywhere? No. Signals not reaching the threshold are blocked completely — no arrow, no Dashboard entry. This differs from other filters (e.g., Direction Filter displays blocked signals as "faded" or with X symbol).

---

Q3: Is Q = 0.35 high or low? Medium. With H1 XAUUSD distribution, threshold 0.35 passes about 47% of signals — a good balance between quantity and quality. Scores below 0.30 are usually very dull patterns.

---

Q4: Can I use different thresholds for different pairs? Yes. Each indicator instance (each chart you attach the indicator to) can have independent thresholds. If you run the indicator on XAUUSD H1 and EURUSD M15 simultaneously — each chart has its own input.

---

Q5: Does Quality Filter relate to other filters? Quality Filter operates independently from Direction Filter, StoplossHunt Filter, Signal Zone Filter, Extreme Zone Filter, etc. A signal must pass all enabled filters to display.

---

Q6: Why is PinBar default (0.55) higher than Inverted Hammer (0.30)? Because each pattern has different score distribution. PinBar's scoring formula creates higher distribution — threshold 0.55 still passes about 10% of best signals. Inverted Hammer has lower distribution so 0.30 is reasonable. Don't compare Q scores between different indicators.

---

7. Usage Tips

Tip 1: Start with filter disabled, then gradually increase threshold

When first using the indicator, disable Quality Filter or set threshold very low (0.20–0.25) to see all signals. Observe for 1–2 weeks, noting which signals win/lose and their Q scores. Then you'll have intuition to choose a threshold that works for you.

---

Tip 2: Combine Quality Filter with other filters in priority order

Quality Filter should be your final filtering layer. Set up market-directional filters first (Direction Filter, Signal Zone Filter) — then use Quality Filter for fine-tuning. Avoid enabling too many filters simultaneously with high threshold — you'll have no signals left.

---

Tip 3: Higher threshold isn't always better

Many traders make the mistake of continuously increasing Quality Filter threshold thinking "take only the best signals." In reality, markets don't wait for you to have perfect patterns — setups with Q = 0.45–0.55 are already excellent and appear with reasonable frequency. Thresholds above 0.65 often lead to "missing too many."

---

Tip 4: Use Dashboard to track Q scores over time

During demo or forward test, pay attention to Q scores of winning and losing signals in the Dashboard. If you see most losing signals have Q below 0.40 — that's a clear sign to increase threshold to 0.40 or 0.45.

---

This guide applies to all indicators in the product suite. Specific Q-score formulas differ between indicators, but the scale (0.0–1.0) and filter operation are identical.

Last updated: v1.15 — Ich Khiem Nguyen