FirstFa
Back to playbooks

Cross-Venue Price Check

Official

Check that the price being analysed agrees with other venues and reference prices, and say when a level comes from one venue only.

Updated Oct 17, 2026 · venues, basis, data-quality

What it is

The same asset can print different prices on different venues, and a tokenized stock can drift from its listing while the listing is closed. Before drawing a level to the cent, this check asks whether the price is one the wider market agrees with.

What to look for

  • The venue. get_candles says which source its candles come from. Name it in the answer.
  • Other venues. When the tools offer the same instrument from another venue (for example a venue choice on the candle tools), read the same timeframe from it and compare the last closes and the recent highs and lows. When only one venue is available, say that the check cannot be done here and keep levels to that venue.
  • Related instruments. compare_assets over the same window for the asset and a close relative (a token and its wrapped form, a tokenized stock and its sector ETF): a gap that opens and closes repeatedly is a basis, not a signal.
  • Session gaps. For tokenized stocks, compare the price at the listing's last close with the price now; a move made while the listing was closed may reverse at the open.
  • Wicks. A wick on one venue that no other venue printed is that venue's liquidity, not the market's level.

How to mark it

  • level for a level both venues agree on, labelled with the venues (OKX + other).
  • zone kind range between the two venues' prices where they disagree, labelled Basis.
  • marker on a single-venue wick, labelled Venue wick.

Scenario logic (analysis, not advice)

  • A scenario built on a level only one venue printed is weaker; say so and widen the invalidation to the agreed level.
  • A persistent gap between venues is context for the reading, not an opportunity to act on.

Invalidation

The check is stale as soon as either venue moves; re-read it with fresh candles when the analysis continues later.

Pitfalls

Do not compare different timeframes or candle closes in different time zones. Differences of a fraction of the spread are noise.