Indicators Description

1. Component Deconstruction¶

This section deconstructs the individual mathematical components used to build the indicator’s logic engine.

A. Cumulative VWAP (Volume-Weighted Average Price)¶

• Specific Configuration:

  • Price Source: hlc3 (the average of the high, low, and close of each bar).

  • Anchor Period: User-selectable between Session, Week, or Month. The calculation resets at the beginning of each new anchor period.

  • Reset Logic: The is_new_anchor variable uses built-in functions to detect the first bar of a new period.

    • Session: Resets on the first bar of the regular trading session or on a new day for intraday timeframes.

    • Week: Resets when weekofyear changes.

    • Month: Resets when month changes.

• Functional Modification: This is not a rolling VWAP. It is a cumulative calculation that aggregates hlc3 * volume and total volume from the anchor point. The formula is raw_vwap = cum_tpvol / cum_vol. This grounds the indicator to a specific starting point, representing the average price paid per share/contract since the session, week, or month began.

B. T3 Moving Average (Tim Tillson’s T3)¶

• Specific Configuration:

  • Source: The raw_vwap calculated above. This is a critical distinction; the T3 is smoothing the VWAP, not the price.

  • Length (len): User-defined integer, default 28. This is the lookback period for the nested EMAs within the T3 formula.

  • Factor (factor): User-defined float, default 0.7. This is the a coefficient in the T3 calculation, controlling the degree of smoothing and lag.

• Functional Modification: The script employs a custom function f_t3 that implements the full T3 formula, which is a form of triple-exponential smoothing designed for superior lag reduction compared to a simple EMA or DEMA.

  • Mathematical Logic: The T3 is a weighted sum of six nested Exponential Moving Averages (EMAs) of the source (raw_vwap).

    • e1 = ema(src, len)

    • e2 = ema(e1, len)

    • ...

    • e6 = ema(e5, len)

  • The final t3 value is calculated as c1*e6 + c2*e5 + c3*e4 + c4*e3, where the coefficients c1 through c4 are derived from the factor (a). This complex weighting scheme results in an exceptionally smooth yet responsive moving average, aiming to filter out market noise from the core VWAP value.

C. ATR (Average True Range)¶

• Specific Configuration:

  • Length (atr_len): User-defined integer, default 14.

  • Source: Standard implementation using the bar’s high, low, and close.

• Functional Modification: None. This is a standard ta.atr() calculation used as a raw measure of market volatility. Its purpose is to provide a dynamic scaling factor for the volatility bands.

D. ATR-Based Volatility Bands¶

• Specific Configuration:

  • Basis: The t3 (T3-smoothed VWAP) line acts as the central axis.

  • Width: The atr value is multiplied by a series of hard-coded constants to define the band levels.

• Functional Modification: These are not statistically derived bands like Bollinger Bands (which use standard deviation). They are volatility-derived.

  • Mathematical Logic: The bands are calculated by adding or subtracting a multiple of the current atr value from the t3 basis line.

  • Multipliers: The script uses four fixed, asymmetrical multipliers: 0.5, 1.0, 1.5, and 2.2. These values define discrete volatility zones around the T3-VWAP mean.

2. Logic Layering & Confluence¶

The script’s intelligence lies in how it layers these components to create a hierarchical filtering system.

• Hierarchical Filtering: The score Variable as a “Gatekeeper”

  1. Primary Filter (Regime Definition): The entire logical flow is governed by the score variable. This variable acts as a state machine defining the market regime.

  2. Calculation: score := t3 > t3[1] ? 1 : t3 < t3[1] ? -1 : nz(score[1])

  3. Function:

     • If the T3-VWAP is rising (t3 > t3[1]), the regime is Bullish (score = 1).

     • If the T3-VWAP is falling (t3 < t3[1]), the regime is Bearish (score = -1).

     • If the T3-VWAP is flat, the score persists its previous state (nz(score[1])).

  4. Gatekeeping Action: All subsequent visual and signal logic is conditional on the score. For example, the lower (bullish) bands are only plotted when score == 1, and the upper (bearish) bands are only plotted when score == -1. This immediately filters out half of the potential signals, forcing the user to only consider trades in the direction of the T3-VWAP’s momentum.

• Interaction Dynamics: Threshold Crosses The engine does not use complex pattern recognition or indicator divergence. Its logic is based entirely on a sequence of Threshold Crosses.

  1. Trend Inception Cross: The first and most important cross is the T3 line crossing its own value from the previous bar (t3 > t3[1]). This flips the score and establishes the dominant trend direction. This is the primary entry trigger for a new trend.

  2. Exhaustion/TP Cross: The secondary crosses are the close price crossing the outermost volatility bands (band_u4 or band_l4). This is used to signal potential trade exits (Take Profit). This dynamic represents price “diverging” from the mean by a statistically significant, volatility-adjusted amount.

3. The Execution Engine¶

This section details the precise boolean logic for trade triggers and exits.

• Boolean Logic: Triggers & Exits

  • Long Entry Signal (long):

    • ta.crossover(t3, t3[1])

    • Translation: Returns true for the single bar where the T3-VWAP value becomes greater than its value on the prior bar. This marks the initiation of a bullish regime.

  • Short Entry Signal (short):

    • ta.crossunder(t3, t3[1])

    • Translation: Returns true for the single bar where the T3-VWAP value becomes less than its value on the prior bar. This marks the initiation of a bearish regime.

  • “Bear TP” Signal (tpbear):

    • show_tp and ta.crossover(close, band_u4) and score == 1

    • Translation: This condition is true ONLY IF:

      1. The score indicates a Bullish trend is active (score == 1).

      2. The close price crosses above the outermost upper band (band_u4).

      3. This is an exhaustion signal for a long position. It indicates that in a strong uptrend, price has accelerated so far above the mean that a pullback or reversal is likely.

  • “Bull TP” Signal (tpbull):

    • show_tp and ta.crossunder(close, band_l4) and score == -1

    • Translation: This condition is true ONLY IF:

      1. The score indicates a Bearish trend is active (score == -1).

      2. The close price crosses below the outermost lower band (band_l4).

      3. This is an exhaustion signal for a short position. It indicates a potential capitulation washout in a downtrend, suggesting a prime moment to take profit.

• Mathematical Constants & Risk Profile

  • T3 Factor (0.7): This value is a well-established default for the T3, offering a robust balance between smoothing and responsiveness. A higher value would increase lag but provide a more stable trend definition, filtering out more whipsaws. A lower value would make the T3 react faster, leading to earlier signals but more potential noise.

  • ATR Multipliers (0.5, 1.0, 1.5, 2.2): These hard-coded values are central to the script’s risk mechanics. The final multiplier, 2.2, sets the threshold for an exhaustion signal.

    • Significance: A 2.2x ATR move from the T3-VWAP is the trigger for a take-profit signal. This value directly controls the strategy’s target. A smaller value (e.g., 1.8) would generate TP signals more frequently and with smaller price moves. A larger value (e.g., 3.0) would require a much stronger, potentially parabolic move to trigger a TP, aiming for larger gains but increasing the risk of price reversing before the target is hit. The 2.2 value is a pre-defined constant that dictates the required magnitude of a trend’s “blow-off” top or bottom to be considered an exit point.