Trade Conviction Engine [AGPro Series]

🧠 Core Idea
How much do all your signals actually agree right now - and is this setup worth acting on?
This script turns six separate market dimensions into one clear conviction score.
📌 Overview / What It Does
Trade Conviction Engine is a multi-factor confluence and decision dashboard. Instead of adding one more isolated signal to your chart, it evaluates six independent dimensions of price action - trend, momentum, market structure, volatility, volume and risk location - and blends them into a single 0-100 conviction score.
Alongside that score it shows a directional bias (Long, Short or Neutral), a clear next-action state, and built-in entry, stop and target reference levels derived from recent structure and your chosen reward-to-risk multiple.
It does NOT predict price, generate buy or sell orders, or automate anything. It is a structured way to read how aligned the current context is, so a discretionary trader can make a faster, more consistent assessment of setup quality and execution readiness.
🎯 Purpose & Design Philosophy
Most traders watch many indicators at once and end up making scattered, inconsistent decisions - one tool says go, another says wait, and there is no shared language between them.
Trade Conviction Engine was built to close that gap. It gives every dimension the same 0-100 scale and combines them transparently, so confluence becomes a number you can read at a glance rather than a feeling. It is designed for discretionary, confluence-based traders who want a disciplined pre-trade checklist condensed into one panel.
⚡ Why This Script Is Different
Most tools focus on a single concept - one oscillator, one zone type, one structure model - and leave the trader to mentally combine everything.
This script does NOT add another standalone signal, and it does NOT hide its logic in a black box.
Instead, it orchestrates six independent modules into one transparent score, and applies a conflict penalty: when the modules disagree, the conviction score is automatically reduced. This prevents the common problem of a "high" reading built on contradictory inputs. Every module score stays visible in the panel, so you always see why conviction is high or low - not just the final number.
⚙️ Methodology
1. Six-Module Evaluation
The engine independently measures Trend, Momentum, Market Structure, Volatility, Volume and Risk Location on each bar.
2. Independent Scoring
Each module is normalised to a 0-100 strength value. The four directional modules (Trend, Momentum, Structure, Volume) also produce a direction; Volatility and Risk Location are direction-neutral and measure execution suitability.
3. Weighted Blend
Module scores are combined using user-adjustable weights into a base conviction value.
4. Conflict Penalty & Bias
The net directional vote sets the bias. The more the directional modules agree with that bias, the more the base conviction is preserved; when they conflict, conviction is reduced.
5. State & Reference Levels
The final conviction maps to a next-action state, and reference entry, stop and target levels are projected from recent swing structure and your reward-to-risk setting.
🗺️ How to Read the Chart
- Panel: the conviction score, bias, action state, each module's score, alignment count, and the entry / stop / target levels.
- On-chart labels: mark meaningful state transitions (Prepare and Ready by default).
- Reference lines: dashed entry, dotted stop, and dotted target for the active bias direction.
- Colors: green tones indicate bullish, pink tones bearish, amber neutral. Module values are colored by their own direction.
🚦 Signals & States
- Stand Aside -> conviction is low; conditions are not aligned.
- Watch -> conviction is building; conditions are developing.
- Prepare -> alignment is strengthening; a setup may be forming.
- Ready -> conviction is high; the measured conditions are well aligned.
- Bias Long / Short / Neutral -> the net direction of the four directional modules.
These describe the current state of the measured conditions. They are not directives to enter or exit.
🔔 Alerts Logic
- Ready (Long) / Ready (Short) -> fires when conviction enters the Ready state in that direction.
- Bias Flip -> fires when the directional bias changes.
- Full Module Alignment -> fires when all four directional modules agree.
Alerts are attention markers that something in the measured context has changed. They are NOT trade instructions.
🧩 Confluence Logic
When Trend, Momentum, Structure and Volume point in the same direction, and Volatility and Risk Location are favorable for execution, the conviction score rises and alignment approaches 4/4. When these inputs disagree, the conflict penalty lowers the score even if individual modules look strong.
📊 When To Use
- Clearly trending markets with defined structure.
- Markets showing healthy volatility expansion.
- Liquid instruments where structure and volume are meaningful.
- As a pre-trade confluence check before applying your own analysis.
⚠️ When NOT To Use
- Low-liquidity or thinly traded instruments.
- Choppy, high-noise conditions with no clear structure.
- Extreme volatility events where levels can be unreliable.
- As a standalone reason to enter or exit a position.
🎛️ Key Inputs
- Module Weights -> change how much each dimension contributes to the blended score.
- Module Settings -> EMA lengths, RSI / ROC lengths, swing bars, ATR lengths, volume averages tune each module's sensitivity.
- State Thresholds -> set where Watch, Prepare and Ready begin, and how decisive a bias must be.
- Risk Location -> stop buffer in ATR and the target reward-to-risk multiple.
- On-Chart Labels -> choose Ready Only, Key States, All States or Off, and set minimum spacing to control density.
- Panel -> theme, location, font size, and show / hide.
🖥️ Interface & Visual Design
The panel uses a single merged header and a clean, color-coded layout so the most important information - conviction, bias and action - is readable at a glance. On-chart labels are spaced and floated away from candles to keep the chart uncluttered, with a clear visual hierarchy between the score, the modules and the reference levels.
🧪 Practical Usage Workflow
1. Read the conviction score and bias in the panel.
2. Check the action state to gauge how aligned conditions are.
3. Review the module breakdown to see which dimensions agree or disagree.
4. Note the entry, stop and target reference levels.
5. Combine this read with your own market analysis and plan before any decision.
🔍 Interpretation Guidelines
A high conviction score means strong agreement among the measured dimensions - not a guarantee of direction. A low alignment count signals conflicting inputs and warrants caution. Treat the score as context, and always interpret it within the broader market picture.
🚫 What This Script Is NOT
- It is NOT a prediction engine.
- It is NOT financial advice.
- It is NOT an automated trading system.
- It does NOT generate guaranteed signals or outcomes.
⚠️ Limitations & Transparency
Module behavior can vary across timeframes and instruments. Structure detection relies on confirmed swing pivots, which introduces a natural delay. Volatility and liquidity conditions affect how levels and scores appear, and changing market regimes can shift how the modules interact. Outputs should always be read in context.
🧠 Market Context Notes
Conviction is most informative when liquidity is sufficient, structure is well defined, and volatility is expanding in a controlled way. In compressed or erratic conditions the same readings carry less weight.
🧾 Use Case Examples
- A trader sees conviction in the Ready state with 4/4 alignment in their bias direction, and uses it as confirmation to study a setup more closely.
- A trader sees conviction drop and alignment fall to 1/4, and treats it as a reason to stand aside and reassess.
🧱 System Philosophy
The engine favors confluence over single signals and transparency over black-box scoring. Every input that drives the final number remains visible, so the trader stays in control of the interpretation.
🔐 Non-Promise Statement
This script makes no promises about future price movement and offers no certainty. It is a measurement and visualization tool only.
📉 Risk Disclosure
Trading involves substantial risk. This script is provided for educational and analytical purposes only and does not constitute financial advice. All trading decisions and their outcomes are the sole responsibility of the user. Past behavior of any indicator does not guarantee future results.
📚 Educational Note
Use Trade Conviction Engine to study how different market dimensions align and diverge over time. Over repeated observation it can help build a more structured, disciplined approach to reading confluence and setup quality.
This update focuses on proof: following each Ready setup to its target or stop and
reporting the historical hit rate and average R.
The core purpose of the script remains unchanged. It still scores six independent
market dimensions, blends them into a 0-100 conviction with a directional bias, a
next-action state, and entry, stop and target reference levels. This release adds an
outcome layer so the track record of Ready setups is visible.
This script continues to function as an analytical and visualization tool. It does
not attempt to predict price direction or provide guaranteed outcomes.
------------------------
What Changed
------------------------
- Outcome proof layer
A new block follows every Ready setup from its entry through to its target or stop,
and reports the pooled Hit Rate and Average R, with the sample size (n). Wins are
measured at the projected target R multiple and losses at the stop, so the result
reflects the plan the panel actually presented.
- Single, honest pool
All Ready setups are pooled into one record rather than split into sub-buckets, which
keeps the statistic stable and easy to read.
- Honest sampling
The outcome rows stay greyed until enough Ready setups have resolved to read them, so
a thin sample is never presented as a strong result.
------------------------
Visual Improvements
------------------------
- Improved chart readability by grouping the outcome statistics into their own panel
block, separated by a divider.
- Reduced visual clutter by leaving the existing module rows, state labels and
reference levels unchanged.
- Refined visual hierarchy so the headline conviction, the six modules, the plan and
the outcome proof each occupy a distinct block.
- Kept the conviction state labels and entry / stop / target lines for continuous
at-a-glance context.
------------------------
Interface & Usability
------------------------
- Added Hit Rate and Avg R rows beneath the reference levels.
- Enhanced overall user experience while keeping panel theme, location and font size
fully adjustable, with the panel shown by default.
- All module, weight, threshold and label settings remain available and unchanged.
------------------------
Behavior Notes
------------------------
This update adds an outcome-tracking layer. The six modules, the blended conviction,
the bias, the state and the reference levels are unchanged, so the live readout
matches the previous version.
The goal is to improve clarity, depth and usability, not to introduce predictive
behavior. The hit rate and average R describe how past Ready setups resolved within
the visible history and are not forecasts. Small samples should be read with care.
------------------------
Limitations Reminder
------------------------
The script remains a rule-based analytical tool.
The outcome statistics are computed only over the Ready setups visible in the loaded
history, and they vary with timeframe, instrument, weights and thresholds. A favorable
reading does not guarantee any individual outcome.
Outputs should always be interpreted within broader market context.
------------------------
Risk Reminder
------------------------
This script is for educational and analytical purposes only.
It does not provide financial advice or guaranteed trading outcomes.
Users remain responsible for their own decisions.
This update focuses on scoring integrity, sample integrity, panel behavior, analytical wording, and completed-bar alert stability.
The core purpose of the script remains unchanged.
This release refines the existing six-module conviction workflow without changing its concept, thresholds, module set, or reference-level design.
This script continues to function as an analytical and visualization tool.
It does not attempt to predict price direction or provide guaranteed outcomes.
------------------------
What Changed
------------------------
• Corrected Risk Location scoring for the active bias.
Risk Location now evaluates the reward-to-risk geometry of the direction selected by the four directional modules. The previous calculation could borrow the more favorable ratio from the opposite direction and overstate conviction.
• Corrected structural stop-side validation.
A confirmed swing is now used as the stop base only when it is on the valid side of the current entry reference. If it is not, the current bar extreme provides the fallback, preserving positive risk distance and correct long/short target direction.
• Corrected historical sample registration.
The outcome layer now registers one sample per confirmed Ready transition or confirmed Ready bias change. A persistent Ready regime can no longer open another sample immediately after the previous sample resolves.
• Corrected panel hide behavior.
The table is cleared on the last bar before it is rebuilt. Turning Show Panel off now removes the existing cells instead of leaving a stale panel visible.
• Confirmed alerts at candle close.
Ready, Bias Flip, and Full Module Alignment alerts now require barstate.isconfirmed.
------------------------
Visual Improvements
------------------------
• Replaced multi-line state labels with compact single-line conviction badges.
• Increased the default historical label spacing from 20 to 32 bars.
This keeps the chart active while reducing badge collisions on compact intraday layouts.
• Standardized chart-facing badge text to premium uppercase casing.
• Improved contrast on bullish, bearish, neutral, and target badges.
• Preserved user-adjustable label spacing and ATR-based displacement away from candles.
• Expanded the merged blue header to the full product name: AG Pro Trade Conviction Engine.
------------------------
Interface & Usability
------------------------
• Replaced the panel's Action row with Context.
• Replaced direct-looking state phrases with analytical context wording such as Ready / Long, Prepare / Short, and Low Conviction.
• Preserved Show Panel, panel location, Dark/Light panel theme, panel font size, label font size, level visibility, and level projection controls.
• Preserved the six module rows, 0-100 conviction score, alignment count, reference levels, Hit Rate, and Avg R.
------------------------
Behavior Notes
------------------------
The directional bias is still derived from Trend, Momentum, Structure, and Volume.
Volatility remains direction-neutral. Risk Location remains a suitability score, but it now measures only the reward-to-risk geometry that belongs to the selected bias.
The outcome statistics remain a conservative, one-position-at-a-time historical summary. If both stop and target are touched within the same bar, stop is evaluated first because intrabar path is unknown.
Context values and alerts are analytical attention markers. They are not entry or exit instructions.
------------------------
Limitations Reminder
------------------------
The script remains a rule-based analytical tool.
Confirmed swing pivots introduce natural delay. Outcome statistics depend on loaded history, timeframe, symbol, thresholds, module weights, and the selected reward-to-risk multiple.
Bar-based historical evaluation cannot reconstruct the exact intrabar path when stop and target are both inside the same candle.
Market conditions such as volatility, liquidity, and timeframe differences may affect how scores, labels, and reference levels appear.
------------------------
Risk Reminder
------------------------
This script is for educational and analytical purposes only.
It does not provide financial advice or guaranteed trading outcomes.
Users remain responsible for their own decisions.
• Refreshed the publication and access routing information.
• Core trade-conviction logic, panel behavior, inputs, alerts, visuals, and outputs remain unchanged.
This update focuses on evidence clarity, chart readability, and premium information hierarchy.
The core purpose of the script remains unchanged.
This release improves how conviction states and their resolved historical outcomes are presented and interpreted.
This script continues to function as an analytical and visualization tool.
It does not attempt to predict price direction or provide guaranteed outcomes.
------------------------
What Changed
------------------------
• Added a dedicated panel question: “IS THIS CONVICTION RELIABLE?”
This creates a clear analytical focus and connects the live conviction score with its supporting evidence.
• Added a 90% confidence floor for the observed hit rate.
The panel now distinguishes the raw observed result from a more conservative lower-bound estimate.
• Added an Evidence Depth classification.
Resolved samples are clearly identified as very early, early, developing, or strong instead of presenting small samples as mature evidence.
• Raised the evidence-ready threshold from 5 to 30 resolved samples.
Outcome colors remain neutral while the sample is still too small for stronger visual emphasis.
• Renamed entry, stop, and target rows as analytical references.
The panel now uses Entry Reference, Risk Reference, and Target Reference for clearer non-directive interpretation.
------------------------
Visual Improvements
------------------------
• Added the standardized product-specific question row beneath the merged blue title
• Reduced all three separator rows to a compact 1% height while preserving section hierarchy
• Increased default state-label spacing from 32 to 48 bars
• Converted state labels to a compact two-line format to reduce horizontal chart coverage
• Refined panel terminology and score formatting for faster first-glance reading
• Preserved adjustable panel theme, location, panel font size, and label font size
------------------------
Interface & Usability
------------------------
• The panel now separates live conviction, module alignment, reference levels, and outcome evidence into a clearer hierarchy
• Small-sample statistics are explicitly framed instead of being visually overstated
• Label density remains user-adjustable through the existing minimum-spacing control
• Dark and light panel themes retain the AGPro contrast-safe color system
------------------------
Behavior Notes
------------------------
This update does not change the six-module conviction model, directional bias logic, state thresholds, reference-level calculations, outcome tracking rules, or alert conditions.
The 90% hit-rate floor and Evidence Depth row are interpretation safeguards applied to the existing resolved-outcome sample.
------------------------
Limitations Reminder
------------------------
The script remains a rule-based analytical tool.
Historical outcome evidence is specific to the loaded symbol, timeframe, available chart history, and current input configuration.
Small samples can vary materially as additional outcomes resolve.
Market conditions such as volatility, liquidity, and timeframe differences may affect how states and reference levels appear.
------------------------
Risk Reminder
------------------------
This script is for educational and analytical purposes only.
It does not provide financial advice, automated trading, or guaranteed outcomes.
Users remain responsible for their own decisions.
This update focuses on live-state ownership, reference-level clarity and premium chart readability.
The six-module conviction engine, scoring logic, conflict penalty, Ready-transition evidence model and alert conditions remain unchanged.
------------------------
What Changed
------------------------
• Added a live conviction card linked directly to the active entry reference. It summarizes state, bias, conviction, directional alignment, core module scores and evidence depth without changing the analytical engine.
• Extended Entry, Stop and Target labels beyond their projected rails and connected each label back to its exact reference price, making visual ownership explicit.
• Renamed historical transition labels as PRIOR CONVICTION and increased candle clearance so past states cannot be mistaken for the current setup.
• Converted the panel Context row into a contrast-safe filled state row and hardened merged header construction for consistent rendering.
------------------------
Visual Improvements
------------------------
• Improved first-glance separation between the live setup, historical states and reference rails.
• Added font-aware spacing to reduce label overlap with candles.
• Applied contrast-aware text colors across state cards and level labels.
------------------------
Behavior Notes
------------------------
This release does not change the six analytical modules, score thresholds, directional bias logic, stop or target calculations, outcome tracking, or alerts.
The script remains a rule-based analytical and visualization tool. It does not predict price direction, automate trading or provide guaranteed outcomes.
------------------------
Risk Reminder
------------------------
Trading involves risk. Outputs must be interpreted within broader market, liquidity, volatility and timeframe context. This script is not financial advice, and users remain responsible for their own decisions.
Invite-only script
Only users approved by the author can access this script. You'll need to request and get permission to use it. This is typically granted after payment. For more details, follow the author's instructions below or contact AGProLabs directly.
TradingView does NOT recommend paying for or using a script unless you fully trust its author and understand how it works. You may also find free, open-source alternatives in our community scripts.
Author's instructions
tradingview.com/script/XJmX3p98-Trading-Suite-AGPro-Series/
🔷 Crypto Setup Planner - Multi-Symbol Planner
tradingview.com/script/LgVRo7Ql-Crypto-Setup-Planner-AGPro-Series/
Disclaimer
Invite-only script
Only users approved by the author can access this script. You'll need to request and get permission to use it. This is typically granted after payment. For more details, follow the author's instructions below or contact AGProLabs directly.
TradingView does NOT recommend paying for or using a script unless you fully trust its author and understand how it works. You may also find free, open-source alternatives in our community scripts.
Author's instructions
tradingview.com/script/XJmX3p98-Trading-Suite-AGPro-Series/
🔷 Crypto Setup Planner - Multi-Symbol Planner
tradingview.com/script/LgVRo7Ql-Crypto-Setup-Planner-AGPro-Series/