Most signal channels post whenever a single indicator flips. SignalBot does the opposite: it demands that several independent pieces of evidence agree before it says anything. Each scan, every pair gets a score, and only setups clearing the threshold are published. This article walks through what actually gets counted.
The five factors
The score adds one point for each aligned factor: trend direction (EMA 9 above or below EMA 21), momentum (MACD histogram sign), RSI positioning (recovering from oversold for buys, rolling over from overbought for sells), price reaction at a nearby support or resistance level, and volume above its recent average. A perfect alignment scores 5; the bot requires at least 3 before it will even consider posting.
The filters that kill weak setups
Clearing the score is necessary but not sufficient. A 4-hour trend filter rejects signals that fight the higher-timeframe direction — a long in a falling market needs the 4H chart to at least be neutral, and in strict mode, bullish. ADX below 20 rejects the trade as chop. A flat EMA slope rejects it as drift. And if a support or resistance level sits between entry and take-profit, the trade is rejected for having no clear path.
Why you see fewer signals than other channels
This stack of filters means most scans end with nothing posted. That is by design. A signal feed's value is not how often it speaks but how much evidence stands behind each message. When SignalBot does post, you know the setup cleared trend, momentum, structure, volatility and volume checks simultaneously — and you can verify every one of them on the TradingView link that ships with the alert.
As always: signals are algorithmic heuristics, not financial advice. They can be wrong. Use them as a starting point for your own analysis, never as a replacement for it.