Live Queries & Streaming

Real-time data feeds, auto-refreshing queries, and streaming market data.

Real-Time Data in the research desk

The research desk is not limited to point-in-time queries. It supports two real-time data paradigms: auto-refreshing queries that re-execute on a configurable interval, and streaming widgets that receive push-based data updates via WebSocket connections. Together, these features turn the research desk into a live monitoring platform alongside its analytical capabilities.

Auto-Refreshing Queries

Any SQL cell in a notebook or widget on a dashboard can be configured to auto-refresh at a set interval. When auto-refresh is enabled, the query re-executes automatically and the results update in place. Any downstream cells (visualization, statistics, counters) that reference the refreshing cell also update automatically.

Configuring Refresh Intervals

Click the clock icon on any SQL cell or dashboard widget to open the refresh configuration panel. Select one of the preset intervals or enter a custom value.

10 seconds	Near real-time. Best for price tickers, funding rate monitors, and liquidation alerts. High credit consumption.
30 seconds	Frequent refresh with moderate cost. Good balance for derivatives monitoring during active trading sessions.
1 minute	The default for most dashboard widgets. Provides timely data without excessive credit usage.
5 minutes	Suitable for metrics that change slowly: z-scores, composite indicators, portfolio P&L.
15 minutes	Low-frequency refresh for summary dashboards and KPI monitors. Minimal credit impact.
1 hour	For daily tracking dashboards where near-real-time data is not necessary.
Custom	Enter any interval in seconds. Minimum is 10 seconds; maximum is 86400 seconds (24 hours).

Credit consumption at short intervals

A SQL query costs 5 credits per execution. At a 10-second refresh interval, that is 30 executions per minute, or 150 credits per 5 minutes for a single widget. Be deliberate about which widgets need sub-minute refresh rates. Most traders find that 1-minute refresh is sufficient for active monitoring.

Smart Refresh Strategies

Not every widget on a dashboard needs the same refresh rate. A well-designed dashboard uses tiered refresh intervals to balance data freshness against credit cost.

Tier 1: Critical (10-30s)

Reserve for 1-2 widgets that display the data you act on immediately: live price, active liquidation events, or real-time funding rate.

Tier 2: Active (1 min)

Use for core monitoring widgets: OI changes, volume spikes, divergence status, alert evaluations.

Tier 3: Context (5-15 min)

Apply to background context widgets: z-score panels, market overview metrics, smart money summaries.

Tier 4: Reference (1 hr+)

Set for slow-moving data: performance analytics, historical comparisons, portfolio allocation breakdowns.

Streaming Widgets

Streaming widgets receive data via persistent WebSocket connections rather than periodic query execution. This provides true real-time updates with sub-second latency and zero per-update credit cost. The initial connection setup costs 5 credits, but subsequent data pushes are free for the duration of the session.

Available Streaming Feeds

Price Ticker	Real-time price updates for any tracked asset. Sub-second latency from aggregated exchange feeds.
Funding Rate Stream	Live funding rate changes across exchanges. Updates every time a new rate is published (typically every 8 hours, with intermediate snapshots).
Liquidation Feed	Real-time liquidation events as they occur across major exchanges. Includes asset, direction, size, and exchange.
Whale Transaction Feed	Live whale transaction alerts as large on-chain transfers are detected. Includes transaction hash, amount, source, and destination.
Trade Stream	Your own trade executions in real time if you have exchange API keys connected. Shows fills as they happen.
Social Sentiment Stream	Real-time shifts in social sentiment scores as new data is ingested from major social platforms and community channels.

Adding a Streaming Widget

Open the widget panel in dashboard edit mode

Navigate to your dashboard, click Edit, and expand the widget panel on the right side.

Select Data Feed widget

Click Data Feed in the widget type list. Drag it onto the canvas grid where you want it placed.

Configure the feed

In the widget configuration panel, select the feed type (price, funding, liquidation, whale, trade, or sentiment). For asset-specific feeds, select which assets to subscribe to. Configure display options: scroll speed, max visible rows, color coding.

Save and view

Save the dashboard. The streaming widget connects immediately and begins displaying real-time data. Events scroll through the widget as they arrive, with the most recent at the top.

Live Notebook Cells

Auto-refresh is not limited to dashboards. Any SQL cell in a notebook can be set to auto-refresh. This is useful when you are iteratively developing a query and want to watch results update as market conditions change. Downstream visualization and statistics cells automatically re-render when their source data refreshes.

To enable auto-refresh on a notebook cell, click the cell's menu icon (three dots) and select Auto-Refresh. Choose your interval and the cell begins refreshing immediately. A pulsing indicator appears in the cell header to show that auto-refresh is active.

Pause auto-refresh when idle

Auto-refresh runs whether or not you are looking at the notebook. If you step away from your desk, pause auto-refresh to conserve credits. Click the pulsing indicator on any cell to toggle refresh on and off. Dashboard auto-refresh automatically pauses when the browser tab is not focused.

Data Freshness Indicators

Every auto-refreshing widget and streaming feed displays a freshness indicator in its header. The indicator shows how long ago the data was last updated and uses color coding to signal staleness.

Green	Data is fresh. Updated within the last refresh interval or within 10 seconds for streaming feeds.
Yellow	Data is slightly stale. Missed one refresh cycle or the stream has been idle for 30+ seconds. May indicate a temporary connectivity issue.
Red	Data is stale. Multiple refresh cycles missed or the stream has been disconnected for 60+ seconds. Click to force a manual refresh or reconnect.

Combining Live and Analytical Data

The most powerful dashboards combine live streaming widgets with analytical query widgets. A typical trading dashboard might include a price ticker stream and liquidation feed (real-time), a funding rate table and OI change chart (1-minute refresh), and a z-score panel with historical context (5-minute refresh). This layered approach gives you both immediate awareness and analytical depth.

Next Steps

Advanced Research desk

Formula builder, backtesting, query scheduling, and automation.

Dashboard Builder

Build the dashboard layouts that host your live widgets.

Notebooks & Cells

Write the queries that power your auto-refreshing widgets.

AI Features

Add AI-powered market summaries to your real-time dashboards.