Methodology & Data Sources

Last updated: April 30, 2026

1. Overview

SpeedSignal scans ~1,900 US equities daily (Russell 3000) at market close (4:15 PM ET) and assigns each ticker a phase — Coiling, Pre-Ignition, Liftoff or Below — plus a 0–100 score and a Conviction level (Low / Medium / High). Higher scores combined with higher Conviction indicate a higher-probability setup.

Our scoring is algorithmic and rules-based. It does not use AI predictions, sentiment analysis, or subjective judgment. Given the same market data, every score is reproducible.

2. The Three Pillars

Each pillar is evaluated as a percentage (0–100%) of conditions met. The final score is the average of the three pillars.

Pillar 1: Base "Stealth Coil"

Measures how tightly a stock is coiling before a potential breakout:

Volatility Contraction — ATR% percentile rank over 252 days (tighter = stronger).
Tightness — Range / ATR ratio. Narrow action = ready.
Squeeze Detection — Bollinger Bands inside Keltner Channels.
Trend Quality — Stage 2 (Close > SMA50 > SMA200).
Pivot Proximity — Distance to nearest resistance.

Pillar 2: Accumulation "Smart Money"

Detects institutional accumulation before the crowd notices:

Relative Strength Alpha — RS line vs SPY at new highs with positive slope.
Money Flow — Chaikin Money Flow (CMF) and On-Balance Volume (OBV) trend.
Volume Dry-Up — Below-average volume during tight range = supply absorbed.
Sector Tailwind — Sector ETF strength vs its SMA50.

Pillar 3: Ignition "Strike"

Identifies the breakout catalyst itself:

Volume Impulse — Unusual volume vs 50-day average, tiered by market cap.
Early Ignition — Range expansion or pressure at pivot.
Candle Quality — Body ratio, close position, up-day confirmation.

3. Phase Classification

Phase	Meaning
Coiling	Tight setup forming. Volatility compressed, price building a base. Watching, not acting yet.
Pre-Ignition	Momentum building. Accumulation visible, ready to move. Earlier-stage candidates.
Liftoff	Breakout active. Price + volume confirming above pivot. Most actionable phase. Sticky once entered (only exits via squeeze re-coil or trend break).
Below	Not in any active setup phase right now.

4. Conviction Level

Inside each phase, we add a Conviction dimension — Low, Medium, or High — based on quantile thresholds of the 0–100 score. Thresholds are admin-configurable; default mode is "Auto Top 5%" (the High band targets the top 5% of scored tickers each session). Conviction is a separate axis from phase: a Coiling ticker can be High Conviction; a Liftoff ticker can be Low Conviction.

5. Risk Flags & Anomalies (informational)

Some tickers carry context flags that do not change the score itself but signal risk worth knowing:

Extended — Price stretched well above SMA50 (chase risk).
Bearish Market Regime — SPY below SMA50 with negative slope.
Elevated VIX — Broad volatility above its threshold.
Pre-Earnings Run — Earnings within ~7 days (event risk).
Momentum-Only / Anomaly — Moves outside the canonical setup (visible in the Anomalies tab).

6. Eligibility Filters

Before scoring, stocks must pass:

Price ≥ $5
Liquidity: ADV20 × Close ≥ $15M
Market Cap ≥ $300M
Free Float > 30%
Not in "dead trend" (Close < SMA200 with negative slope)
No excessive gap down
Earnings shield (buffers around earnings dates)

7. Data Sources

Data Type	Source	Update Frequency
Price & Volume	Exchange data via TradingView	Daily at close
Fundamentals	FactSet via TradingView	Quarterly
Earnings Dates	Exchange filings	As filed
Market Regime	SPY, VIX, Sector ETFs	Daily
Company Info	syminfo (TradingView)	Static

8. Limitations

Scores reflect end-of-day data — intraday moves are not captured
Historical scores may differ from live scores due to data revisions
Earnings data availability varies by ticker
Small-cap stocks ($300M–$2B) use wider thresholds due to different market dynamics
The system is optimized for daily timeframe breakout detection

9. Disclaimer

This methodology description is provided for transparency and educational purposes. Scores are not investment advice or recommendations. Past performance of high-scoring stocks does not guarantee future results. Always conduct your own research and consult a qualified financial advisor.