Indicators Description

1. Component Deconstruction¶

This section deconstructs the primary mathematical components of the “Algo Trend System TG” script.

Supertrend Indicator¶

• Specific Configuration:

  • Function: ta.supertrend()

  • ATR Lookback Period (atrPeriod): 10. This defines the period for the underlying Average True Range (ATR) calculation, which measures volatility.

  • ATR Multiplier (factor): 3.0. The calculated ATR value is multiplied by this factor to determine the offset of the trend bands from the price’s median.

  • Price Source: The ta.supertrend function internally uses (high + low) / 2 as the median price from which the ATR-based offset is calculated.

• Functional Modification:

  • This is a standard, unmodified implementation of the Supertrend indicator. Its primary role is not as a standalone signal generator but as a tactical, volatility-adjusted trigger mechanism. The script does not use the plotted supertrend line for its logic, but rather the internal direction series returned by the function. This series holds a value of -1 for a bullish trend and 1 for a bearish trend.

Exponential Moving Average (EMA) Cloud¶

• Specific Configuration:

  • Function: ta.ema(source, length)

  • Fast EMA (emaF): A 50-period EMA calculated from the close price.

  • Slow EMA (emaS): A 150-period EMA calculated from the close price.

• Functional Modification:

  • The two EMAs are not used as independent crossover signals. Instead, their relative position (emaF > emaS or emaF < emaS) is used to establish a persistent “state” or “regime.” This transforms the EMAs from lagging indicators into a single, binary regime filter that defines the permissible trade direction over a macro timeframe. The visual fill between the two lines reinforces this concept of a continuous trend environment.

2. Logic Layering & Confluence¶

The script’s intelligence derives from its hierarchical filtering process, where signals must pass through sequential logical gates to be considered valid.

• Hierarchical Filtering: The engine employs a two-tier filtering architecture:

  1. Gatekeeper (Regime Filter): The EMA Cloud (50/150) serves as the primary filter. It establishes the strategic directional bias. A buy signal is only considered if the market is in a bullish regime (emaF > emaS), and a sell signal is only considered in a bearish regime (emaF < emaS). This layer is designed to filter out all trades that run counter to the dominant, long-term trend, significantly reducing the number of potential signals.

  2. Trigger (Tactical Filter): The Supertrend (10, 3.0) acts as the secondary, more sensitive filter. It is not used to define the overall trend but to identify a specific event: the resumption of momentum after a pullback.

• Interaction Dynamics: The core logic is built on Confluence following a specific narrative: “a pullback to the mean and its subsequent rejection.”

  • For a Long Trade: The system does not trigger when the Supertrend is simply bullish. It waits for a precise sequence:

    1. The EMA Cloud must be bullish (emaF > emaS).

    2. The price must have pulled back sufficiently to flip the more sensitive Supertrend indicator to a bearish state.

    3. The trigger event is the resumption of strength, identified by the exact bar on which the Supertrend flips back to a bullish state. This is captured by ta.crossunder(direction, 0).

  • This layering ensures the script only engages when short-term momentum (Supertrend) realigns with the long-term trend (EMA Cloud), aiming to capture the continuation move.

• Ancillary Filters:

  • Temporal Filter: An optional time filter (inAllowedTime) acts as an additional gate, allowing signals only within specified trading sessions. This reduces noise by focusing on periods of expected higher liquidity or volatility.

  • State Filter: A state machine (inTrade, tradeDirection) prevents the system from taking a new trade in the same direction while one is already active, effectively enforcing a “one trade at a time” rule per direction.

3. The Execution Engine¶

This section details the precise boolean logic for trade execution and the mathematical constants governing risk management.

Trigger Conditions¶

• Buy Signal (buySignal): A true value is returned if the following four conditions are met on the same bar:

  1. ta.crossunder(direction, 0): The internal Supertrend state variable crosses from 1 (bearish) to -1 (bullish).

  2. emaF > emaS: The 50-period EMA is above the 150-period EMA.

  3. inAllowedTime == true: The bar’s time is within a valid session (if enabled).

  4. tradeDirection != 1: The system is not currently in a long trade.

• Sell Signal (sellSignal): A true value is returned if the following four conditions are met on the same bar:

  1. ta.crossover(direction, 0): The internal Supertrend state variable crosses from -1 (bullish) to 1 (bearish).

  2. emaF < emaS: The 50-period EMA is below the 150-period EMA.

  3. inAllowedTime == true: The bar’s time is within a valid session (if enabled).

  4. tradeDirection != -1: The system is not currently in a short trade.

Exit Conditions & Mathematical Constants¶

The script utilizes a fixed-target exit mechanism, not a dynamic or indicator-based one. All profit and loss levels are calculated from the close of the signal bar.

• Mathematical Constants:

  • Stop Loss (sl_dist): A fixed point/dollar value (default: 20.0). This value is subtracted from a long entry’s close or added to a short entry’s close. The risk per trade is therefore static in terms of price points, not percentage or volatility.

  • Take Profits (tp1_dist, tp2_dist, tp3_dist): Fixed point/dollar values (defaults: 10.0, 20.0, 30.0). These are added to a long entry’s close or subtracted from a short entry’s close.

• Risk-to-Reward Profile:

  • The script’s R:R is entirely dependent on the user-defined inputs. With default settings, the profile is as follows:

    • R:R to TP1: 10:20 = 0.5:1

    • R:R to TP2: 20:20 = 1:1

    • R:R to TP3: 30:20 = 1.5:1

• Trade Invalidation (for Statistical Tracking):

  • The internal dashboard logic tracks a single “win” condition based on the closing_tp input.

  • A trade is marked as a Loss if price touches the stopLoss level (low <= stopLoss for longs; high >= stopLoss for shorts).

  • A trade is marked as a Win if price touches the user-selected final take profit level (high >= finalTPTarget for longs; low <= finalTPTarget for shorts).

  • This is a “first-touch” model; the simulation does not account for partial profits or complex trade management, treating the entire position as closed once one of these two levels is breached.