Sign inCreate account
← Strategy library

Relative Activity Filterthe cheapest improvement most intraday systems can make

Evidence: ModerateIntermediateOrder flow · Filter

Not a strategy — a gate. Compute each bar's activity relative to the same bar's own recent average, and refuse every entry that arrives while activity is below average.

The reasoning is not subtle. A break that happens on twice the usual activity has participants behind it. The identical break on a third of the usual activity is often one feed's quotes drifting through a level, and it retraces.

Read the data page in this book before trading anything here. What your platform calls volume on a currency pair is a tick count, and every page in this family depends on knowing what that does and does not measure.

Category
Order flow · Filter
Timeframe
M5 to H1
Suitable pairs
Any · works on indices with genuine volume too, and better there
Session
Any — the filter is what tells you whether the session is real
Indicators / tools
Tick volume (or real volume on futures) · its 20-bar average · the strategy you are filtering

Valid setup

OF-07 · RELATIVE ACTIVITY FILTER — VALID SETUP GBPUSD · M15 · the same level broken twice — once on below-average activity and once on well above LONG 1.2665 1.2685 1.2706 1.2726 1.2747 1.2767 1.2788 2R 1R ENTRY = the break that passed the gate THE LEVEL — approached twice STOP = 1.2 × ATR(14) below the pullback low 1 2 3 1.2718 1.2703 1.2689 relative activity — the gate opens above 1.5 (here: bar range ÷ average range) 1.5 gate average ① First break of the level — activity below average ② Skipped by the gate, and it fails within four bars ③ The signal was recorded, not taken ④ Second break, activity well above the gate — taken ⑤ Stop and target come from the underlying strategy ⑥ Test it with the gate off. That comparison is the point

Exact rules

Entry

  • Compute relative activity = this bar's tick volume ÷ the average of the last twenty bars.
  • Gate: entries allowed only when relative activity on the signal bar is above 1.5. Below that, the signal is recorded and skipped.
  • Compare like with like where you can — the same hour on previous days rather than the trailing twenty bars — because activity has a strong time-of-day shape that a trailing average confuses with a signal.
  • Apply it to the signal bar, not to the bars after it. A filter that needs the next bar to confirm is a delay, not a filter.

Stop loss

  • Unchanged from the underlying strategy. The filter decides whether to enter, not where the risk goes.
  • Expect the filtered version to trade considerably less. That is the entire mechanism.
  • Risk as the underlying strategy specifies.

Take profit / exit

  • Unchanged from the underlying strategy.
  • Measure it properly: run the same strategy with and without the gate over the same period and compare. This is one of the easiest things in the library to test honestly.
  • If the filter does not improve the result, drop it. A filter that does nothing is worse than no filter, because it makes you confident.

Build Relative Activity Filter

Reading this page is free and always will be. Taking it is what costs — the complete MQL5 for the rules above, yours to keep, permanently. Five of them cost $10 and they never expire.

Sign inWhat it costs

More from Order flow & volume

London 16:00 Fix Flowmonth-end rebalancingSigned Order Flow Continuationtrade with the tape, not against itVolume-Confirmed Liquidity Sweepthe spike is the evidence

Trading carries risk and most retail accounts lose money. Nothing on this page is financial advice or a recommendation to trade. The chart is an illustration of the rules on constructed candles — it is not a record of a trade and not a backtest.