Order flow surge
Marks one-bar volume bursts that close higher with positive money flow; they come often.
OrderFlowSurge marks a bar when four things hold together: its volume is over 1.5 times the average of the last window bars, including itself; money flow over that window is above 0.15; the close is above the previous close; and a combined surge score is above 2. Money flow weights each bar's volume by where its close sits in its high-low range, from -1 at the low to +1 at the high. It reads only bar price and volume, with no order book or institutional data, despite the name.
How to read it
- Each marker is one bar where all the surge conditions hold on that bar's volume and price.
- Markers are common on SPY one-minute bars, about 5% of bars and roughly 38 a day in the sample, so a single marker says little.
- A cluster of markers in a short stretch means repeated volume bursts with rising closes.
- The sample does not show whether markers lead to price moves; check that on your own chart.
What it outputs
- Surge
- A marker on each bar where the surge conditions all hold. The underlying score is above 2 wherever it appears, and nothing is drawn elsewhere.
When it fires
When this bar's volume is over 1.5 times the average of the last window bars, money flow over that window is above 0.15, the close is above the previous close, and the surge score is above 2. It means a heavy bar closed higher while money flow looked positive. It does not show who traded.
Settings
- window
- Length of the volume and money-flow average. Short windows let one spiky bar set the average and fire often; at 1 it never fires, since a bar always equals its own average. Long windows need a heavier bar to beat the average and slow the money-flow reading.
Pitfalls
- The name overstates it: it fires on about 5% of SPY one-minute bars, roughly 38 a day, so it detects ordinary volume bursts rather than institutional flow.
- The money-flow term divides by the current bar's high-low range in dollars, so the 0.15 gate and the score move with bar width and price level. Treat it as a rough filter, not a calibrated money-flow reading.
- Only the default window was measured on the sample, so how other windows change the marker count is not shown here.