TickerNote's whole premise is that you should be able to check the homework before you trust the call. So here is the process in plain language — what goes into a signal, how a score becomes a BUY / HOLD / SELL, and, just as importantly, what this method does not claim to do.
Read this first. Everything TickerNote publishes — the morning report scores, the on-demand verdicts, the backtest, and the public paper portfolio — is research and education, not investment advice, and all performance figures are paper / simulated. Our own out-of-sample testing has not demonstrated a durable, net-of-cost edge over the market.Any tilt we've measured so far is small, gross of costs, and unproven across market regimes. We publish the method so you can judge it for yourself — not because we've proven it beats the market.
The daily morning report combines up to eight families of signal into a single integer score. Each factor is computed deterministically from primary-source data — there is no black-box model and no discretionary override. A higher score means more independent signals are lining up constructively; it is a way to rank attention, not a prediction.
Where price sits relative to its 50-day and 200-day moving averages. Above both is a constructive uptrend; below both is a downtrend.
The 14-period Relative Strength Index. Oversold readings (below 30) can flag mean-reversion setups; overbought readings (above 70) flag near-term pullback risk.
Distance from the 52-week high and low. Trading near the high is relative strength; sitting near the low is persistent weakness until a base forms.
Short-window price change (e.g. 30-day) to characterise whether a name is extended, basing, or in a heavy drawdown.
Company financials from primary filings — used in the daily report to sanity-check that a technically strong name isn't fundamentally broken.
Aggregated analyst recommendations from Finnhub, used as one corroborating input — never as a standalone reason to act.
Options-chain context and Form 4 insider transactions from SEC EDGAR, as additional confirmation or caution flags.
The broader regime (e.g. VIX, rates) and sector context, so a single name is read against the tape rather than in isolation.
The integer score maps to a label. In the on-demand analyzer, for example, a score of ≥2 reads as BUY, 1 as Accumulate, 0 as Hold, −1 as Trim, and below that as Avoid. These are mechanical outputs of the published formula describing how a textbook setup would be framed — not personal recommendations.
When a setup scores as a BUY, we illustrate textbook levels using the same formula everywhere — the morning report, the on-demand analyzer, the backtest, and the paper trader all share it:
These are illustrative levels for how a disciplined entry would be framed — not a recommendation for any individual, and not a guarantee that any level will be reached.
The live report uses all eight factors, but four of them — fundamentals, analyst consensus, options flow, and sentiment — can't be reliably reconstructed for every ticker on every past day. So both our public backtest preview and the monthly Python backtest deliberately use a technicals-only approximation of the score. Treat any historical trade list as indicative of what the method flags, not a perfect replay of the live engine.
Backtests also assume no transaction costs, slippage, taxes, or dividends, and are computed after the fact. Hypothetical performance has well-known limitations and does not reflect actual trading. Real-world results would be modestly lower.
Because a backtest can always be over-fit in hindsight, we also run the signals forward in public. A paper portfolio executes the report's BUY/SELL signals every weekday — drawdowns and losing trades included — and shows the gap versus the S&P 500. We don't reset or cherry-pick that record. You can watch it on the live track record.
Replay the technicals-only score on stocks you actually follow. No signup.
The signals, traded forward in public — wins, losses, and the gap vs SPY.
TickerNote is research and educational software, not investment advice. Signals are computed descriptions of price action and corroborating data, not predictions. All performance shown anywhere in the product is paper / simulated; past performance does not guarantee future results. See the full disclaimer.