Home › Methodology

How Our AI Signals Work

A transparent look at the engine behind TradeSignal's daily stock picks.

Signal Generation Process

1Nightly Scan

Every night after market close, our AI engine scans 190+ stocks across the US (S&P 500) and Swedish (OMX Stockholm) markets.

2Multi-Factor Scoring

Each stock is scored from 1-10 using a combination of technical indicators, chart patterns, momentum signals, volume analysis, and sector rotation data. The scoring engine evaluates dozens of factors simultaneously.

3Signal Filtering

Only stocks scoring above a minimum threshold become signals. This high-conviction filter ensures quality over quantity — typically 10-20 signals per day out of 190+ stocks scanned.

4Entry & Exit Levels

Each signal includes a calculated entry price, stop-loss level, and profit targets based on ATR (Average True Range) and support/resistance analysis.

What We Analyze

Backtested Performance

Our signals are backtested across multiple time periods and market conditions. We test in-sample (2019-2025), out-of-sample (2010-2018), and stress-test during bear markets (2022). All backtests include realistic transaction costs (slippage + commissions).

Important: Backtested performance is not indicative of future results. All trading involves risk, and you should never invest money you cannot afford to lose. Our signals are informational tools, not financial advice. Always do your own research before making any investment decisions.

Data Sources

All our market data comes from publicly available sources including Yahoo Finance for price data, SPDR sector ETFs for rotation analysis, and SEC filings for insider activity. See our full data sources page for details.

Transparency Commitment

We believe in transparency. Our signal methodology is rule-based and systematic — no discretionary overrides, no cherry-picking. Every signal that passes our threshold is published, regardless of whether we think it will work. This eliminates survivorship bias in our track record.