What this chart measures
Persistence is the fraction of eligible sampled observations in which a bucket meets the wall rule. A resting order can be cancelled before a trade reaches it.
Select the asset, exchange and timeframe before comparing readings. Each venue response is normalized to the selected spot market. The chart shows completed candles and labels snapshot age; missing or stale observations remain visibly unavailable.
Use the other evidence without conflating it
Current order-book depth represents displayed resting liquidity. Candle volume is completed activity over the selected interval. CMF and OBV are disclosed price-and-volume proxies. They are useful for describing different parts of an observation but do not identify whale holdings or guarantee a direction.
The detailed book below the chart lets you check quantities and quote notionals behind the overlay. Compare a proposed size against available displayed depth, then account separately for fees, price changes, cancellations and fragmented venue liquidity.
Controls and reproducibility
Change the visible range, zoom or pan, choose a secondary study and switch the price scale. Export the available data or a chart SVG to preserve a research record. Session liquidity data must first be collected or imported; chart history is not manufactured from price candles.
Distinguish repeated buckets from persistent orders
A persistent liquidity cluster means a qualifying price bucket appeared in consecutive retrieved observations. The book response does not provide a durable identity for every order, so a repeated bucket cannot establish that one order or one participant remained there. Treat persistence as a property of the sampled display, not a verified account of trader behavior.
The result depends on the bucket width, cluster threshold and sampling window. A wider bucket can combine nearby prices; a stricter threshold can remove a cluster without any underlying cancellation. Keep those settings fixed during a comparison. If a request fails or a long gap occurs, there is no observation of what happened in between, even if the next snapshot looks similar.
Use repeated clusters to choose a more precise research question: was usable depth consistently available near a proposed execution price during the observed session? The follow-up is a quantity and cost calculation, followed by a comparison with real fills if a trade later occurs. Neither duration nor size proves support, resistance, common ownership or an intention to execute.
The session display positions observations by their recorded times and leaves long sampling gaps blank. A persistence filter changes which repeated clusters remain visible; it does not create missing observations. Pausing stops collection in this page, so the resulting session is neither continuous market surveillance nor a historical archive.
Read the source and freshness labels before using a result. Public feeds can be delayed, blocked or unavailable by market, region or browser. Pause stops this page's feed collection. No chart observation guarantees a fill, identifies an order's owner or establishes future returns.