Guide

How to read Top Confluence Picks

The AI Assistant tab synthesizes all six analysis views — Price/Volume, Candlestick, Screener, Momentum, Volatility, and Volume — into a single confluence score and verdict per stock. The top ten ranked stocks appear as Top Confluence Picks, with an optional AI watchlist brief and chat for follow-up questions.

Confluence score

Each analyzed stock receives a numeric confluence score that reflects how many bullish (or bearish) signals align across the six tabs. Higher scores mean broader agreement; negative scores mean bearish alignment dominates.

The score is computed deterministically from cached unified analysis — no extra per-stock AI calls. The same indicator data powers the individual Screener, Momentum, and Volume tabs.

What each tab contributes

TabAffects score?Role in confluence
Price/VolumeYesAI signal (BUY / HOLD / SELL) and confidence — up to ±3 points each.
CandlestickYesAI signal and confidence — same weight as Price/Volume.
ScreenerYesStrong-buy filters (MACD bullish, RSI oversold, near support, above SMA-20) add points; strong-sell filters subtract. ±2 per matched filter.
MomentumYesComposite momentum score: +2 if ≥70, +1 if ≥55, penalties if weak (<45).
VolumeYesSurge/Elevated volume adds points; Dry Up subtracts. Rising OBV adds +1; falling OBV −1.
VolatilityNo (score)ATR % is shown for context and used as a ranking tiebreaker — lower volatility ranks higher when scores tie. It does not add or subtract from the confluence score.

Verdict badges

After scoring, each stock gets a plain-language verdict:

VerdictTypical meaning
Strong BuyHigh confluence score (≥12), no opposing AI signals, no BUY vs SELL conflict.
BuyModerate bullish alignment (score ≥6) without SELL signals on both AI tabs.
HoldMixed or neutral alignment — not clearly bullish or bearish.
AvoidBearish confluence or both AI tabs show SELL.
ConflictingPrice/Volume and Candlestick disagree (e.g. BUY vs SELL). Review both tabs before acting.
Insufficient DataAnalysis still loading or incomplete for this symbol.

Top Confluence Picks prefer stocks with Strong Buy verdicts, then conflict-free Buy verdicts. If fewer than ten qualify, remaining slots fill from the next highest scores.

OHLCV period

Top Confluence Picks are computed from unified analysis over a selectable OHLCV window: 15D, 30D (default), or 60D trading days. Use the period buttons in the Top Confluence Picks header on the AI Assistant tab.

The selected period is shared across all six analysis tabs (Price/Volume, Candlestick, Screener, Momentum, Volatility, and Volume). Changing it re-analyzes your watchlist and may change confluence scores, verdicts, and rankings.

Shorter windows (15D) emphasize recent price action and patterns; longer windows (60D) incorporate more history. Compare periods when you want to see whether a pick holds up across different time horizons.

How picks are ranked

Stocks are ordered by:

  1. Confluence score — highest first.
  2. Volatility (ATR %) — lowest first when scores tie (calmer stocks rank above more volatile ones at the same score).
  3. Ticker symbol — alphabetical tiebreaker.

Live prices in the table refresh when you open the AI Assistant tab, so quote data reflects current market snapshots rather than prices frozen at analysis time.

Symbols classified as exchange-traded funds show a small ETF badge next to the ticker. Fair value triangulation does not apply to ETFs, so the Fair value column shows an em dash (—) instead of bear/base/bull scenario links for those rows.

Conflicts to review

Even high-scoring stocks can show warnings when tabs disagree. Common conflict flags:

  • Price/Volume BUY but Candlestick SELL (or the reverse)
  • Bullish AI signals with weak momentum (score below 45)
  • Bullish AI signals with drying volume
  • Bullish AI signals but bearish screener filters matched

Conflicts do not always disqualify a stock from the top ten, but they appear in the orange Conflicts to review section so you can investigate before relying on the pick.

Watchlist brief & chat

The Watchlist brief is a short prose summary of your watchlist generated from the same confluence JSON sent to the model. It highlights top picks and conflicts using only your actual data — no invented tickers or numbers.

Ask about your analysis lets you chat in plain language (e.g. “Compare my top two picks”). Answers are grounded in your watchlist context only.

If OpenAI is not configured, the brief falls back to a deterministic text summary built from the same scores and verdicts.

Important caveats

  • Not financial advice. Confluence picks are informational synthesis tools, not trade recommendations.
  • Technical only. Scores do not include fundamentals, earnings, news, or macro events.
  • Shared cache. All tabs use the same 12-hour unified analysis cache; refresh analysis on other tabs to update confluence.
  • Top ten is relative. Picks are the best alignments within your watchlist, not a market-wide screen.
  • AI summaries can err. Always verify against the individual tab tables before making decisions.

Access Top Confluence Picks from the Analysis page under the AI Assistant tab.