Using the Technical Screener: 5 Filters That Actually Matter
The technical screener ships with over 40 parameters. Most traders use three or four — and the wrong three or four. This guide identifies the filters that consistently produce high-signal results in trending markets and explains why the others tend to generate noise.
Why more filters don't mean more signal
More filters narrow the output but don't improve its quality. Combining ten loosely correlated conditions usually produces a smaller, equally noisy result set. Signal comes from filters that are both individually meaningful and mutually reinforcing — each one measuring a different dimension of the same underlying condition.
The 5 filters
These five parameters consistently outperform on signal quality in trending markets. Each measures a different aspect of market structure, which is why they complement rather than duplicate each other.
- RSI(14) between 40–60: pulled back from overbought, resetting in an uptrend
- 50-day MA slope positive: confirms the trend direction is intact
- Volume ratio > 1.2: the move has above-average participation
- Price above 200-day MA: long-term trend filter, removes bear market stocks
- ATR(14) percentile 40–70: enough volatility to move, not so much it's chaotic
How to combine them
These five work together because they measure different things: momentum reversion (RSI), trend direction (50d MA), participation (volume), macro trend (200d MA), volatility regime (ATR). When all five align, you're looking at a trending stock in a consolidation phase with rising participation — historically a high-probability setup in bull market regimes.
What doesn't work and why
Pattern-and-candlestick filters, gap filters, and price-level breakout filters all have much lower signal-to-noise ratios because they're too recent to confirm a trend and too volatile to use without multi-session confirmation. They work as secondary confirmation — not as the primary filter driving your screener output.
Key Takeaway
Save this five-filter combination as a named preset. Run it weekly, track output versus results, and resist the urge to add more parameters when the list feels too short. Fewer, better filters consistently outperform longer, noisier ones.
More in Guides