Explosive momentum breakout
Marks one-minute bars that jump 1%+ on heavy volume and close near their 20-bar high. Never fired on a month of SPY at defaults.
The indicator reads each one-minute close and volume and checks three things on the same bar. The close must be more than 1% above the close one bar earlier, the bar's volume must exceed 1.5 times the average of the last 20 bars (the current bar included), and the close must sit within 0.5% of the highest close of the last 20 bars. Only when all three hold does it output a number, the one-bar gain multiplied by the volume ratio, and it outputs nothing on every other bar. The 1%, 1.5 and 0.5% thresholds are fixed in the code, so only the three windows can be tuned.
How to read it
- Markers appear only on candles where all three tests pass. Most candles carry none, and that is the normal state.
- Each marker is a sudden one-minute push of 1% or more on heavy volume that closes near the 20-bar high.
- A marker shows what happened on that bar. It is not a forecast of the next moves.
What it outputs
- Explosive breakout
- Fires on a one-minute bar that passes all three tests. The value is the one-bar gain times the volume ratio, so a larger number means a stronger setup.
When it fires
Fires when a bar closes more than 1% above the bar before it, on volume above 1.5 times its 20-bar average, and within 0.5% of the 20-bar closing high. It marks a sudden, volume-backed push to a local high.
Settings
- roc_period
- Bars back the gain is measured over, where 1 means the previous close. Raising it lets a slower 1% climb over more bars count, so more bars tend to pass. At 1, the minimum, only a single-minute jump of 1% qualifies, which is the strictest setting.
- vol_window
- Bars in the volume average, current bar included. Raising it usually lifts the volume ratio on a given spike, so more bars pass. At 1 the average is the bar itself, so the ratio is always 1 and the indicator never fires.
- resistance_window
- Bars in the closing-high lookback, current bar included. Raising it makes the close need to sit near a higher peak from further back, so fewer bars pass. At 1 the window holds only the current close, so the closing-high test always passes and drops out.
Pitfalls
- It never fired on the sample: 0 events in 20,160 one-minute SPY bars from 2026-09-01 to 2026-10-01 at defaults, so a blank chart is the expected result. The try presets were not backtested, so some may stay blank too.
- The docstring says highest high, but the code compares closes. The test means a close near the 20-bar closing high, not a break above an earlier resistance level.
- The 1%, 1.5 and 0.5% thresholds are fixed in the code and are not parameters, so only the three windows can be tuned.