> ## Documentation Index
> Fetch the complete documentation index at: https://docs.speedsignal.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Data & coverage
> Where Speed Signal AI's data comes from, when it updates, which 3,000+ US tickers are eligible, and why a given ticker may not appear in the scan.
Speed Signal AI scans **3,000+ US tickers** every trading day: the Russell 3000, S\&P 500, and Nasdaq 100, across NYSE, NASDAQ, and AMEX. This page covers where the data comes from, when it refreshes, and what makes a ticker eligible for the scan.
## When the data updates
Scores update **once per day, after the US market closes**. Everything you see reflects the most recent completed trading session, not live intraday prices. The only live element is the candlestick chart on a ticker's detail page.
Because everything reflects the most recent completed session, the numbers you see in the morning describe yesterday's close. They do not move during the trading day.
## Where the data comes from
| 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 context | SPY, VIX, and sector benchmark ETFs | Daily |
| Company reference | Exchange/company reference data (TradingView) | As company info changes |
The scoring itself is **deterministic and rules-based** and runs on the same market data every time. It uses no AI, sentiment, or predictions.
## What you see
Most of the eligibility thresholds below are adjustable from the [Signal Board](/dashboard/signal-board) in its **Advanced Filters** drawer, where each threshold has inline help.
## Eligibility filters
Before scoring, a ticker must pass a set of coverage filters so the board stays focused on tradable US equities. The values below are the defaults:
| Filter | Default |
| ------------------- | -------------------------------------------------------------------------------------------- |
| **Price** | ≥ \$2 |
| **Liquidity** | a minimum average daily dollar-volume |
| **Market Cap** | ≥ \$300M |
| **Free Float** | > 30% |
| **Trend** | not in a clearly established downtrend |
| **Gap** | no excessive gap down |
| **Earnings buffer** | scoring pauses within a window around confirmed earnings dates, to reduce event-driven noise |
## Why isn't my ticker here?
If a name you follow does not appear in the scan, it usually falls into one of these buckets:
* **It's outside the covered universe.** The scan covers the Russell 3000, S\&P 500, and Nasdaq 100 on US exchanges. Tickers outside those indexes (including most non-US listings, OTC names, and funds) are not scanned.
* **It didn't pass an eligibility filter.** A price under $2, a market cap under $300M, thin liquidity, a low free float, a clear downtrend, or a large gap down will keep a ticker out of the scan for that day.
* **It's inside an earnings buffer.** Scoring pauses in the window around a confirmed earnings date, so a name can drop out temporarily around its report.
* **It's eligible but in the Below phase.** It passed the filters but does not currently meet any active setup criteria, so it may not surface in the views you were looking at. In the [Signal Board](/dashboard/signal-board) Advanced Filters drawer, enable the **No Setup** status pill (the label the Below phase uses in the dashboard tables) to include these names.
Many filters are adjustable in the Signal Board's Advanced Filters drawer. Loosening a threshold (for example, lowering the market-cap floor) can bring more tickers into the list.
## Limitations
* Numbers reflect **end-of-day** data; intraday moves are not captured.
* Historical numbers may differ from live ones because of **data revisions**.
* **Earnings-date availability varies** by ticker.
* **Small-cap stocks can behave differently** from larger companies; the board treats that context separately where relevant.
* The model is **optimized for daily-timeframe breakout detection**, not other styles or timeframes.
## Next steps
The three pillars that turn this data into the 0–100 number.
Adjust the eligibility thresholds and filter the full list.
What the phase labels and Conviction levels mean.
Plain-language definitions of every label you see in the app.