STRUCTURA CORE is a multi-factor quantitative framework that measures the statistical "quality" of a potential trade setup before execution. Rather than generating directional signals from price patterns alone, the system asks a prior question: is the market currently exhibiting persistent, measurable structure that makes trend-following statistically viable? Only when the answer is affirmative — and quantified above a calibrated threshold — does the system produce a tradeable signal.
At its core, STRUCTURA produces two composite scores:
Composite of entropy dynamics, Hurst differential, and a 3-bar structural integrity check. RES ≥ 0.75 defines a "Strong" regime — the only regime in which the system enters trades.
Real-time measure of market self-similarity, geometric confluence, and directional efficiency. Acts as a gating condition and as an exit trigger when it degrades below 0.40.
The Core Thesis: Most retail and many professional traders apply fixed-parameter indicators to markets that cycle between trending, mean-reverting, and random regimes. This mismatch — a trending indicator applied in a mean-reverting regime — is the primary source of avoidable losses. STRUCTURA detects these regime states continuously and only trades when the current regime is quantifiably aligned with the trading strategy.
The system is built on seven statistically grounded components organised into two composite scores. All parameters are calibrated for H4/Daily timeframes with default lookback N=100. The v2.9 Python engine is available for independent validation.
| Metric | Method | Role | Composite | Reference |
|---|---|---|---|---|
| H_L / H_S | DFA Hurst (12-scale, R²≥0.90) | Trend persistence; directional bias | RES → HDI_norm (40%) | Peng et al. 1994 |
| Shannon Entropy | Freedman-Diaconis binning (≤40 bins) | Regime randomness detection | RES → ECR_norm via R3 (25%) | Shannon 1948 |
| EDS | Bipower variation vs. realised variance | Jump/diffusion energy transition | RES → EDS (35%) | Barndorff-Nielsen & Shephard 2004 |
| Yang-Zhang Vol | 3-component OHLC estimator | Position sizing; vol regime | Sizing scalar (0.5×–2.0×) | Yang & Zhang 2000 |
| H_stable | Cross-window Hurst variance | Estimate stability / noise filter | SII (40%) | Original |
| G_conf | Fibonacci / Gann proximity (11 levels) | Price-level structural confluence | SII (30%) | Heuristic — see §5 |
| E_dir | Efficiency ratio (net / gross path) | Directional efficiency | SII (30%) | Perry Kaufman 1995 |
Invalidation Gates — A trade signal is suppressed entirely if any of four conditions are active:
The v2.9 engine incorporates all findings from a full independent audit conducted prior to this document. Key execution properties are listed below. The full audited Python source is available to qualified counterparties upon request.
| Parameter | Specification |
|---|---|
| Entry timing | Signal fires on bar N → entry at bar N+1 open price. No lookahead bias. |
| Exit timing | All exits (SL, TP, degradation) execute at the subsequent bar's open (gap model). If exit_at_stop=True, SL fills at the stop price (guaranteed-fill model). |
| Exit priority | SL/TP checked before signal degradation. Exit reason attribution is accurate. |
| Position sizing | Risk = 1% of equity per trade. Scaled by YZ volatility ratio (target_vol / current_vol), capped at 0.5×–2.0×. Hard cap: 5% notional of equity. |
| Stop / target | Stop: rolling N-bar low (long) or high (short) minus spread + slippage. Target: 2R from entry. Minimum stop distance: 5 pips. |
| Costs | 1 pip spread + 0.5 pip slippage per side, deducted from PnL and R-multiple denominator. |
| Cooldown | 3-bar mandatory cooldown after any exit. Prevents overtrading at regime transitions. |
| Sharpe test | Block bootstrap (5,000 iterations, block=5 bars) on active-bar returns only. p-value < 0.05 required for statistical significance. |
Regime Breakdown — The table below confirms the system's primary value: performance concentrates strongly in "Trending / Low Entropy" setups. This is not luck — it is the direct consequence of the RES and SII filters doing their job.
| Regime at Entry | Trades | Win Rate | Avg R | Total PnL |
|---|---|---|---|---|
| Trending / Low Entropy | 61 | 57.4% | +1.08R | +++ |
| Trending / Mid Entropy | 74 | 49.3% | +0.61R | ++ |
| Random / Mid Entropy | 38 | 36.8% | +0.12R | +/− |
| Trending / High Entropy | 14 | 35.7% | −0.08R | − |
The following assessment is written for a technically sophisticated audience. It deliberately surfaces risks and unresolved questions alongside genuine differentiators. Obfuscating the weaknesses of a quantitative system is both intellectually dishonest and ultimately counterproductive.
The direct answer is: yes, conditionally — and more genuinely than the majority of TradingView indicators at this price tier.
STRUCTURA CORE is not a generic indicator. The vast majority of indicators sold on TradingView are cosmetic re-codings of RSI, MACD, Bollinger Bands, or moving average crossovers. STRUCTURA is architecturally different: it is a regime filter, not a signal generator. That is a meaningful distinction, and the underlying statistics (DFA Hurst, Yang-Zhang vol, bipower variation) are defensible in academic literature.
However, "better than generic" is not the same as "allocator-ready." The gap between the two requires specific work.
TradingView indicator access, the v2.9 Python engine, raw OHLCV backtest data, and this full methodology document are available upon request to qualified institutional counterparties. For retail access, purchase directly via the main site.