# FinsburyPark2_updown — Extreme Volatility + Simultaneous Volume Spike Detector_NO REPAINT (MT4)
Hi all,
Sharing a custom indicator I've been refining — **FinsburyPark2_updown**. It's built around a fairly simple statistical idea: real reversal moves tend to happen when an extreme price move and an extreme volume move occur **on the same bar**, not just one or the other. Most "spike" indicators only look at price; this one requires both to line up before it calls anything a real signal.
**How it works:**
1. **Price spike (ATR-based):** the bar's range (high–low) is compared to the current ATR. If the range is a large multiple of ATR (default 2x), it's flagged as an abnormal bar.
2. **Direction via wick rejection:** rather than just using candle color, it checks which side got rejected — a long lower wick (≥50% of the bar's range by default) signals a bullish rejection → Buy; a long upper wick signals bearish rejection → Sell.
3. **Volume confirmation (z-score, not just "above average"):** it computes the mean and standard deviation of volume over a lookback window, then requires the spike bar's volume to be a set number of standard deviations above that mean (default 2.0σ) — a proper statistical outlier test rather than a fixed multiplier, so it adapts to each instrument's own volume behavior.
4. **Optional extra filters:** Bollinger Bands excess (price must actually be outside the bands), RSI extreme, EMA trend alignment, session/time filter, and a cooldown between signals to avoid clustering.
**Visuals:**
- **Main chart:** small arrows mark every detected signal — **cyan = Buy**, **red = Sell**.
- **Separate subwindow:** a volume histogram that stays flat except on genuine statistical spikes (color-coded cyan/red by direction), with a mean line and a dotted statistical threshold line for reference. On volume-confirmed spikes, a small stack of markers plus a star appear right at the peak — a quick visual "this is where price and volume lined up."
Everything runs through native MT4 buffers/objects — no DLLs.
Happy to share the full parameter list or discuss the volume z-score logic further if anyone's working on something similar. Feedback and suggestions welcome, especially from anyone who's tested volume+volatility confluence approaches before.
