Methodology

1. Market

Current price, gap, and session state are read from live or most-recent available quote data. All prices are 2 decimals, IST timezone. If the market is closed or data is stale, the system shows DATA UNAVAILABLE or MARKET CLOSED.

2. Structure

CPR (Central Pivot Range), PDH (Previous Day High), and PDL (Previous Day Low) are derived from the previous trading day's OHLC only. Current day's candles never affect own CPR. CPR is structural context — never a standalone signal.

3. Gap

GAP_POINTS = TODAY OPEN − PREVIOUS CLOSE. GAP_PERCENT = (OPEN − PDC) / PDC × 100. Opening location relative to CPR is classified as ABOVE_CPR, INSIDE_CPR, or BELOW_CPR using only the opening price and D-1 CPR.

4. Price Action

Breakouts, breakdowns, holds, rejections, and reclaim are identified from completed 5-minute candles relative to CPR levels. A confirmed breakout requires price action beyond CPR sustained over completed candles. The current structure and confirmed trade trigger are displayed separately.

5. Confirmation

Confirmation uses completed 5-minute candles only. An incomplete candle (e.g., 09:15 bar before 09:20) is NEVER treated as confirmation. The 09:15 print is an observation, not an entry.

6. Decision

DECISION is one of: TRADE (fully qualified and approved), WAIT (setup developing, confirmation incomplete), NO TRADE TODAY (no qualified setup), DATA UNAVAILABLE (missing data), MARKET CLOSED (outside session). The system does NOT manufacture trades. WAIT and NO TRADE are valid outcomes.

7. Timing

Signal → Trigger → Confirmation → Actionable → Entry. These are recorded separately and never collapsed. Entry cannot occur before confirmation. Every decision explains its timing.

8. Risk

Invalidation is defined by the setup: for above-CPR scenarios, loss of CPR bottom invalidates; for below-CPR scenarios, breach of CPR top invalidates. Each setup shows its invalidation level where available.

9. Research

Point-in-time (PIT) replay validates that no future data was used at any historical timestamp. Future-mutation tests confirm that mutating data after time T does not change any result at or before T.

10. Limitations

• Only approximately 59 clean 5m sessions currently available per instrument
• Insufficient for predictive conclusions (≥250 sessions recommended)
• Historical options data unavailable
• Credit-spread backtesting unavailable
• Historical volume data unavailable (zero for all candles)
• No performance claims (win rate, profit factor, etc.) are made