Picture the one-minute chart of a major pair at 12:59:30, a few seconds before a big release. The spread is a pip or so, candles are tiny, and everyone is staring at the same number on the calendar.
At 12:59:45 the spread starts to creep, maybe half a pip wider. At 12:59:55 it jumps. Quotes flicker, the depth that looked comfortable a minute earlier is gone, and some platforms begin updating prices far less often. Then the release hits, and for a second or two the price is mostly a guess made by whoever is still quoting.
The reason is not mysterious. Anyone providing liquidity takes on risk every time they quote two sides. If the number lands far from expectations, every quote left standing becomes a loss. The sensible move is to widen the quote or pull it entirely until the first reaction is visible. When many providers do that at the same moment, the market looks like it vanished.
What reaches a retail platform is a mix of this. Spreads widen, fills slow down, and in some cases the broker disables new orders for a few seconds. The chart can look smooth afterward because candles only record prices that were actually quoted, which hides how little was tradable in the middle.
The practical point is that the price visible in those seconds is not the price available. Two people clicking at the same moment can get very different results, and neither one made a mistake. It also explains why the move right after a release sometimes overshoots and snaps back. Liquidity returns, quotes get re-centered, and the spike that looked like a trend turns out to have been a vacuum.
Anyone who has watched a raw spread feed or depth display through this window sees it clearly, and the timing differs by pair and by release.
How early do you see spreads start widening on the pairs you trade, and is it different for NFP than for a lower tier release?
What the last 30 seconds before a release actually look like
-
Shadow Trader
- Posts: 5438
- Joined: Wed Sep 09, 2026 5:19 pm