Market Regime & No-Trade Decision Agent

Market Regime & No-Trade Decision Agent is a premium trading-environment classifier and strategy-suitability gate designed to solve a problem that conventional signal agents often ignore:

A valid trading strategy does not perform equally well in every market regime.

Instead of beginning with:

"Should I buy or sell?"

the agent begins with:

"What kind of market exists right now?"

and then asks:

"Is the user's strategy actually designed for this environment?"

This creates an independent market-condition layer between strategy signals and execution.

The agent classifies the market into one primary regime:

STRONG BULLISH TREND

WEAK BULLISH TREND

STRONG BEARISH TREND

WEAK BEARISH TREND

RANGE

COMPRESSION

BREAKOUT ENVIRONMENT

EXTREME VOLATILITY

ABNORMAL EVENT REGIME

It can also produce secondary market-state attributes including:

Trend Direction Trend Strength Trend Quality Directional Efficiency Range Quality Compression Intensity Breakout Pressure Breakout Quality False-Breakout Risk Volatility State Volatility Percentile Liquidity Quality Spread State Gap State Session Quality Higher-Timeframe Alignment Transition Risk Event Risk

The central workflow is:

Validate Market Data → Identify Instrument and Timeframe → Measure Trend Direction → Measure Trend Strength → Measure Trend Quality → Detect Range Structure → Detect Compression → Evaluate Breakout Conditions → Measure Volatility → Detect Abnormal Event Behavior → Determine Primary Regime → Assign Confidence → Load Strategy Profile → Compare Strategy Requirements → Review Historical Regime Performance → Evaluate Execution Conditions → Apply No-Trade Rules → Return Final Suitability Decision

The agent does not assume that trending markets are universally desirable or ranges are universally undesirable.

The correct decision depends on the user's strategy.

For example:

A trend-following system may prefer:

STRONG BULLISH TREND

STRONG BEARISH TREND

while avoiding:

RANGE

COMPRESSION

A mean-reversion system may prefer:

RANGE

while avoiding:

STRONG BULLISH TREND

STRONG BEARISH TREND

BREAKOUT ENVIRONMENT

A breakout system may treat:

COMPRESSION

as a setup environment but return:

WAIT

until expansion and structural confirmation occur.

A scalping system may tolerate multiple structural regimes but become unsuitable when:

Spread Expands Liquidity Deteriorates Volatility Becomes Extreme

The agent therefore builds a strategy-specific Regime Suitability Profile.

A strategy profile may define:

Optimal Regimes Acceptable Regimes Reduced-Risk Regimes Wait Regimes Unsuitable Regimes Minimum Volatility Maximum Volatility Trend Requirements Range Requirements Breakout Requirements Session Requirements Spread Limits Liquidity Requirements Event Restrictions Higher-Timeframe Alignment Requirements

The final result is always exactly one of four operational states:

TRADE NORMAL

The current environment matches the supplied strategy profile and the strategy may operate under its normal user-defined risk policy.

TRADE NORMAL does not mean BUY.

TRADE NORMAL does not mean SELL.

It means only:

The market regime is compatible with the strategy when that strategy generates its own valid entry.

REDUCE RISK

The strategy remains compatible with the current market, but secondary conditions are degraded.

Possible causes include:

Weak Trend Quality Elevated Transition Risk Higher-Timeframe Conflict Volatility Near the Strategy's Maximum Tested Range Reduced Liquidity Wider Spread Early Breakout Conditions Historical Performance That Is Positive but Weaker in This Regime

The agent does not invent a reduction percentage.

If the user defines:

Reduced Risk = 50% of Normal Size

the agent can enforce that rule.

Otherwise it reports:

REDUCE RISK

without manufacturing position-sizing instructions.

WAIT

The market is temporarily uncertain, transitional, unconfirmed, or operationally unsuitable.

Examples include:

Compression Before Breakout Confirmation Insufficient Trend Confirmation Conflicting Timeframes Active Event Blackout Extreme Volatility That Has Not Normalized Temporarily Wide Spread Poor Liquidity Low Regime Confidence Incomplete Market Data High Transition Risk

Every WAIT output should include:

Re-evaluate When:

followed by explicit conditions.

Examples:

Confirmed Range Break Volatility Returns to Tested Range Higher Timeframes Align Event Blackout Expires Spread Normalizes Trend Strength Recovers Breakout Retest Holds

STRATEGY NOT SUITABLE

The detected market structure is fundamentally incompatible with the strategy's supplied design or validated historical performance.

Examples include:

Trend Strategy in Persistent Range Range Mean-Reversion Strategy in Strong Breakout Breakout Strategy During Persistent False-Breakout Range Scalping Strategy During Extreme Volatility Beyond Its Tested Range Strategy with Negative Historical Expectancy in the Current Regime Strategy Explicitly Restricted from Abnormal Event Conditions

This is a structural mismatch rather than temporary uncertainty.

The agent uses multiple sources of trend evidence rather than relying on a single indicator.

Possible trend inputs include:

Price Relative to Moving Averages Moving-Average Alignment Moving-Average Slope Higher Highs and Higher Lows Lower Highs and Lower Lows Directional Movement

ADX

Regression Slope Momentum Persistence Directional Efficiency Pullback Depth Higher-Timeframe Structure

It separates:

Trend Direction

from:

Trend Strength

A market can be bullish while still exhibiting weak trend quality.

This distinction is particularly important for momentum and breakout systems.

STRONG BULLISH TREND may be supported by:

Price Above Major Trend References Rising Moving Averages Higher Highs and Higher Lows Persistent Positive Momentum High Directional Efficiency Shallow or Orderly Pullbacks Strong Follow-Through Bullish Higher-Timeframe Alignment

WEAK BULLISH TREND may contain:

Positive Direction Bullish Bias Moderate or Weak Slope Frequent Overlap Deep Pullbacks Limited Follow-Through Inconsistent Momentum

STRONG BEARISH TREND and WEAK BEARISH TREND are evaluated symmetrically.

Range detection is treated as a structural classification rather than simply "low ADX."

Possible range evidence includes:

Repeated Boundary Rejection Flattening Trend References High Candle Overlap Low Directional Efficiency Repeated Mean Reversion to Midpoint Failed Breakout Attempts Bounded Price Movement

The agent can report:

Range High Range Low Range Midpoint Range Width Range Width in ATR Time in Range Breakout Attempts False-Breakout Count

Range quality can be classified as:

CLEAN RANGE

WIDE VOLATILE RANGE

NARROW RANGE

UNSTABLE RANGE

TRANSITIONAL RANGE

The skill explicitly distinguishes:

Weak Trend

from:

Range with Directional Drift

because these conditions can produce materially different strategy behavior.

Compression is identified through volatility contraction.

Possible evidence includes:

Low ATR Percentile Declining Realized Volatility Low Band Width True-Range Contraction Narrowing Price Structure Inside-Bar Clusters Declining Participation Lower Highs and Higher Lows

Compression does not automatically imply that a breakout will occur.

The agent can report:

Compression Detected Compression Duration ATR Percentile Breakout Readiness Breakout Pressure Direction Unknown

until evidence supports an actual break.

Breakout environments are evaluated structurally.

The agent can distinguish:

PRE-BREAKOUT

INITIAL BREAKOUT

CONFIRMED BREAKOUT

FAILED BREAKOUT

POST-BREAKOUT EXPANSION

Possible breakout evidence includes:

Established Prior Range Boundary Break Close Outside Structure ATR Expansion Volume Expansion Follow-Through Reduced Immediate Rejection Successful Retest

Breakout quality can be classified:

HIGH QUALITY

MODERATE

LOW

FALSE-BREAKOUT RISK

One candle outside a range is not automatically classified as a confirmed breakout.

Repeated boundary failures may instead support:

RANGE

with:

HIGH FALSE-BREAKOUT RISK

The agent also identifies EXTREME VOLATILITY.

This state should not be confused with merely above-average volatility.

Possible evidence includes:

Extreme ATR Percentile Extreme Realized Volatility Abnormally Large Candles Large Gaps Spread Expansion Liquidity Deterioration Rapid Acceleration Abnormal Slippage Range Expansion Beyond Strategy's Tested Conditions

The relevant comparison is not simply:

"Volatility is high."

It is:

"Volatility is materially outside this strategy's normal operating range."

ABNORMAL EVENT REGIME is reserved for environments dominated by unusual events or structural dislocations.

Examples include:

Central-Bank Announcements

CPI

NFP

Geopolitical Shocks Flash Crashes Exchange Outages Trading Halts Unexpected Policy Announcements Liquidity Vacuums Extreme Gaps Instrument-Specific Events Data-Feed Anomalies

The agent can distinguish:

Primary Regime:

ABNORMAL EVENT REGIME

Underlying Structure:

STRONG BULLISH TREND

when event conditions temporarily dominate an otherwise trending market.

Event restrictions come from the user's strategy or risk policy.

The agent does not invent blackout periods.

For example:

No New Trades: 5 Minutes Before CPI 10 Minutes After CPI

can be enforced if supplied.

The agent supports multi-timeframe analysis.

A market may be:

5m:

RANGE

1h:

STRONG BULLISH TREND

This may represent:

Intraday Consolidation Inside Higher-Timeframe Uptrend

The output can therefore include:

Execution Timeframe Context Timeframe Higher Timeframe Alignment

with:

ALIGNED

MIXED

CONFLICTED

A short-term strategy and long-term strategy may interpret the same multi-timeframe stack differently.

The agent evaluates timeframe conflicts through the supplied strategy profile.

For example:

1m: Strong Bearish

15m: Range

1h: Strong Bullish

A swing trend strategy requiring multi-timeframe alignment may receive:

WAIT

The agent calculates Regime Confidence separately from Strategy Suitability.

Example:

Regime Confidence:

HIGH

Strategy Suitability:

LOW

This prevents a high-confidence classification from being mistaken for a high-confidence trading opportunity.

Confidence can be:

HIGH

MEDIUM

LOW

based on:

Indicator Agreement Market-Structure Agreement Timeframe Agreement Data Freshness Regime Persistence Transition Risk Event Uncertainty

Transition Risk can be:

LOW

MODERATE

HIGH

Possible regime transitions include:

COMPRESSION → BREAKOUT

BREAKOUT → STRONG TREND

BREAKOUT → FAILED BREAKOUT → RANGE

STRONG TREND → WEAK TREND

WEAK TREND → RANGE

RANGE → COMPRESSION

RANGE → BREAKOUT

EXTREME VOLATILITY → NORMALIZATION

ABNORMAL EVENT → EXTREME VOLATILITY → TREND / RANGE

The agent can produce transition-aware WAIT decisions.

Example:

Current:

COMPRESSION

Breakout Pressure:

HIGH

Strategy: Breakout

Final:

WAIT

Re-evaluate When: Confirmed range break plus volatility expansion.

The agent can apply hysteresis rules supplied by the user to reduce excessive regime switching.

Example:

Enter Trend Regime: ADX > 25 for 3 bars

Leave Trend Regime: ADX < 20 for 3 bars

The agent does not invent these thresholds.

The skill supports both rule-based and externally supplied statistical regime models.

Possible inputs include:

Hidden Markov Model States Clustering Results Volatility-State Models Machine-Learning Classifier Labels Probabilistic Regime Outputs

The agent can map those outputs into its standard regime taxonomy.

It never fabricates probabilities.

The strongest strategy-suitability evidence comes from the user's own historical performance segmented by regime.

Recommended metrics include:

Trades Expectancy Profit Factor Maximum Drawdown Win Rate Average Trade Consecutive Losses

Example:

STRONG BULLISH TREND:

Expectancy +0.18R

WEAK BULLISH TREND:

+0.05R

RANGE:

-0.07R

COMPRESSION:

-0.02R

BREAKOUT ENVIRONMENT:

+0.21R

A trend strategy with negative historical range expectancy can therefore receive:

STRATEGY NOT SUITABLE

during a high-confidence range.

The agent does not infer strong conclusions from tiny regime samples.

Insufficient historical evidence is labeled:

DATA REQUIRED

The skill can also compare:

Long-Term Regime Performance

with:

Recent Regime Performance

to detect possible strategy-regime drift.

Possible rolling views include:

Last 20 Regime Trades Last 50 Regime Trades Rolling 3 Months Rolling 6 Months

A strategy that historically worked in ranges but recently deteriorated may receive a warning rather than automatic TRADE NORMAL.

The agent includes a volatility-suitability layer.

A strategy can define:

Minimum Volatility Preferred Volatility Maximum Volatility

Example user policy:

ATR Percentile 20–80: Normal

80–95: Reduced

Above 95: Wait

Below 20: Unsuitable

The agent can enforce this exact profile without generalizing it to other strategies.

The same applies to:

Trend Strength Range Width Spread Liquidity Breakout Quality Session Higher-Timeframe Alignment

The skill includes an execution-quality gate.

A market can be:

STRONG BULLISH TREND

and:

Highly Compatible with Trend Strategy

yet still produce:

WAIT

because:

Spread Is Too Wide Liquidity Is Poor Slippage Is Abnormal Event Risk Is Active

This prevents market-structure analysis from ignoring execution reality.

The agent supports session-specific strategies.

Examples include:

US Regular Trading Hours London Session New York Session Asia Session London/New York Overlap

A suitable structural regime outside the strategy's permitted session can still be rejected.

The skill is designed to integrate with other specialized trading-control agents.

Recommended architecture:

Market Regime & No-Trade Decision Agent → Strategy Signal Engine → AI Trading Risk & Position Sizing Guardian → Prop Firm Rule Compliance Trading Agent → Execution Layer

Each component answers a separate question.

Market Regime Agent:

Is this environment appropriate for the strategy?

Signal Engine:

Does a valid setup exist?

Risk Guardian:

Is the account permitted to take this risk and at what maximum size?

Prop Firm Compliance Agent:

Does the proposed action comply with account rules?

Execution Layer:

Can the authorized order be submitted?

This separation reduces the risk of one trading agent approving its own assumptions.

The skill can produce multi-market dashboards.

Example:

Instrument Regime Confidence Strategy Suitability Decision

This is useful when scanning:

NQ ES Gold

EURUSD

BTC

for different strategy environments.

The agent can also maintain historical regime attribution.

For each trade, it can record:

Entry Regime Exit Regime Regime Transition During Trade

This helps evaluate whether a strategy's performance is genuinely associated with the intended market conditions.

The agent can evaluate a regime filter itself.

Compare:

Base Strategy

versus:

Base Strategy + Regime Gate

Metrics include:

Net Return Profit Factor Expectancy Maximum Drawdown Trade Count Exposure Missed Winners Avoided Losses

The agent explicitly recognizes that a regime filter can itself become overfit.

Warning signs include:

Many Optimized Thresholds Narrow Threshold Values Too Many Regime States Small Samples Regime Labels Defined After Seeing Results Filters That Remove Mostly Historical Losers Repeated Tuning on the Same Dataset

Regime-filter validation should therefore include:

Out-of-Sample Testing Walk-Forward Testing Market Holdouts Threshold Perturbation Feature Ablation Parameter-Stability Analysis

The agent can test threshold sensitivity.

If:

ADX > 24

produces excellent results but:

ADX > 23

and

ADX > 25

collapse, the filter may be historically over-specialized.

The skill can also perform feature ablation.

Examples:

Trend Only Trend + Volatility Trend + Range Filter Trend + Volatility + Event Filter

This helps identify whether each additional gate contributes stable value.

The skill treats no-trade decisions as first-class outputs.

WAIT and STRATEGY NOT SUITABLE are not failures to produce an answer.

They are explicit trading-control decisions designed to prevent a strategy from operating outside its intended domain.

The objective is not to maximize trading frequency.

The objective is to ensure that the strategy operates only when the current market environment is compatible with its validated behavior.