Analyze any finance handle.
Type an X handle. We fetch their posts, classify recommendations with Claude, pull price data, and compute the real returns. One analysis takes about 90 seconds and consumes one quota slot.
See who's actually beating the market.
Every account is scored on two horizons — 1M and 3M — short enough that the call has to be right now, not "someday." Receipts on public stock calls, marked to real prices, for returns you can use this quarter. Not financial advice.
If you wanted to invest today, where?
Every open call, collapsed per handle and ticker — with how many times it's been called (conviction) and how it's done since the first call. Flip between by handle and by ticker to see where the best track records are clustering right now.
Follow everyone, tracked daily.
The daily-rebalanced backtest up top — the realized equity curve of holding the leaderboard's composition, benchmarked against SPY over the same open→open windows. Below it, the live composition that portfolio holds right now.
Illustrative backtest + model weights of an impersonal strategy — not investment advice, not a recommendation, and not a solicitation to transact. Past performance does not predict future results. Disclaimer.
Each day Pₜ takes the leaderboard composition; the realized return is the 1-day open-to-open P&L. Cumulative NAV chains those from a 1.0 base.
One portfolio aggregated across every handle's open positions — weight driven by leaderboard score and cross-handle consensus. An illustrative target weight per (ticker, direction).
When independent callers agree, listen.
One rule, found in this site's own data and graded on fixed windows: when two or more tracked handles call the same US stock long within seven days, that consensus has beaten both solo calls and SPY. Below — the evidence, what the rule says today, and a virtual book that has been acting on it every session.
Research notes & a simulated portfolio — not investment advice, not a recommendation, and not a solicitation to transact. Past performance does not predict future results. Disclaimer.