Guide · Updated 2026-11-01

An AI trade setup score is not a win probability

Published by Intelligec · Editorial and evidence policy

An AI trade setup score summarizes an assessment of a trade under a particular model. Without published definitions and validation, a high score does not tell you the chance of winning. Intelligec describes scoring the initial manual MT4/MT5 setup once, then adjusting SL/TP or closing the trade according to its direction judgment.

A number needs a definition

A score may rank setups, summarize a judgment or place observations into categories. Those meanings are not interchangeable. Before treating it as actionable evidence, ask what the scale represents, which outcome it refers to and whether the same meaning holds across instruments and market conditions.

Intelligec has not published a validated scoring rubric, decision thresholds or calibration results. Do not assume its score uses any particular indicator or checklist. A confident explanation is not a substitute for measured performance.

Why 80 does not automatically mean 80%

If a hypothetical tool displays 80 out of 100, that alone does not mean eight of ten similar trades will win. A probability interpretation requires validation against clearly defined outcomes. Winning itself needs a definition: reaching a target before a stop differs from having a positive return after costs at a chosen horizon.

Calibration asks whether predicted probabilities match observed frequencies across enough comparable cases. Even a well-calibrated win probability would not alone determine profitability. Average win, average loss, costs and exposure matter. No such validation is asserted for Intelligec.

Hypothetical example, not live results

Imagine two unrelated scoring systems both assign 80 to the same buy. One ranks trend alignment; the other summarizes a written assessment. Their identical numbers need not have identical meanings. Suppose the trade later loses: that single result cannot establish either system's overall quality.

Likewise, a low-scored trade that wins does not prove all scoring useless. Evaluate a predefined sample and method, including rejected trades where feasible, rather than selecting impressive screenshots. This example does not reproduce Intelligec outputs or demonstrate its effectiveness.

A practical score-review checklist

This is general education, not the actual Intelligec model:

Ask what the score represents and what it explicitly does not represent.
Separate a ranking claim from a probability or profitability claim.
Look for sample size, dates, costs and out-of-sample testing in any performance evidence.
Check whether results cover the instruments and conditions you trade.
Record the original assessment before seeing the outcome, to reduce hindsight bias.

What follows the Intelligec assessment

The described workflow starts with your manual MT4 or MT5 entry. Intelligec detects it and uses market data to score the initial setup once. If direction looks right, it adjusts stop-loss and take-profit. If judged wrong, it closes the trade. Closing is part of the core behavior, not an optional suggestion.

It does not keep monitoring or repeatedly rescore the position. Journaling is secondary. This description is not a guarantee of response time or execution reliability. The AI may close a future winner or retain a future loser.

Keep the score separate from your risk budget

Choose maximum per-trade loss, aggregate exposure and a daily loss limit before entry. Do not double size because a score looks strong. Recalculate exposure if protection levels change, and remember that slippage or gaps can exceed planned losses. Inspect actual position status after automated actions.

Risk sizing is a separate calculation: Position Sizer's documentation, for example, describes sizing from entry, stop-loss, risk tolerance and account size. That is not evidence validating an AI score.

Intelligec is not publicly available yet and the site is a waitlist, with no release date. Explore setup analysis and the described copilot workflow before deciding whether its automatic actions match your needs.

Confidence does not replace the risk calculation

An AI trade confidence score is an assessment output, not a reason to change your loss budget. Read the worked XAUUSD tick-risk example to separate a score from stop distance, volume, commission and execution assumptions. A high setup score does not turn a hypothetical reward-to-risk calculation into a validated probability of profit.

Questions

Is a score a probability of winning?

Not without defined meaning and validation. A score of 80 need not mean an 80% chance of winning.

Does a higher score justify a larger position?

Not by itself. Keep position size within your predefined loss and exposure limits.

Does Intelligec update the score throughout a trade?

No. Its described workflow scores the initial setup once and acts, without ongoing monitoring.

Join the waitlist

Trading involves substantial risk. Stops do not guarantee execution price. AI analysis and automatic closing may be inaccurate or delayed and are not financial advice. Intelligec is in early access.