Research

Every signal on this site is something we have tested against market history before relying on it. This section explains how our tools work and publishes the studies behind them, including what didn’t hold up.

How our tools work

Studies

Our backtests and studies of market indicators, newest first. Each one lists its data, method and limits.

  • Fast Isn’t Enough: How We Build the GSR Select List November 5, 2026
    Most fast growers don’t stay fast. Here is the one number we lean on most to find the ones that can, and the five tests every stock must pass to make our GSR Select list.
  • Twenty Forecasts for the Year Ahead October 30, 2026
    Where 20 of Wall Street’s best-known strategists and investors see the S&P 500 over the next 12 months: most expect higher prices, many expect a dip first, and a vocal few see a bubble.
  • Do Strong Industry Groups Make Better Breakouts? October 11, 2026
    We tested 21,000+ breakouts against the strength, momentum and breadth of their industry groups. A strong group added little once the stock’s own strength was counted. A fast-climbing group helped a bit, and the stock’s own RS mattered most.
  • Testing Our Market Bias Flag: Ten Years of Growth Breakouts October 10, 2026
    Our Market Bias flag is now set by a tested rule: the Nasdaq’s trend plus distribution days. On red-flag days, growth-stock breakouts worked about half as often and a 10% Nasdaq drop followed nearly half the time.
  • Why We Watch Volume, and Whose Volume We Count October 10, 2026
    Big funds can’t hide their selling: it shows up as down days on heavy volume. Here’s how we use our 20-Day A/D Score to manage risk, and why it counts NYSE volume for the S&P 500 and QQQ volume for the Nasdaq.
  • Breadth Beats Volume: Testing Eight Gauges of Institutional Selling October 10, 2026
    We backtested eight measures of selling pressure against nine years of S&P 500 drawdowns. The share of stocks below their 50-day average was the most reliable warning; volume-based measures, including our own distribution-day count, were weaker and less consistent.
  • Do Distribution Days Predict the S&P 500? A Five-Year Backtest October 2, 2026
    We tested our 20-day accumulation/distribution rules against five years of SPY data. A high distribution count flagged elevated drawdown risk, but it did not predict lower prices.

Market basics

For information only, not investment advice. See our Disclaimer.