Skip to main content
Track Volume confirmation
14 / 40
Library

1996issue C031-7

A volume-gated moving-average trend combination

A moving-average-crossover supplies the trend-state switch, and volume-confirmation is added so new positions are not taken from the price average alone. The same 20-session close-through-average rule is written as an always-in-market trend-following system and as an abstention version that waits for session volume above the prior 10-session average.

  • The combination uses a moving-average-crossover as the trend-state switch and volume-confirmation as an entry filter, so signals are not taken from price averages alone.
  • The always-in-market baseline stays long or short after every close through a 20-session average. The volume-gated version opens a new position only when that crossover occurs with session volume above the average of the prior 10 sessions.
  • Exits remain a close back through the price average alone. A crossover without volume-confirmation produces abstention rather than an immediate reverse.
  • A trend-following engine can be paired with a volume, momentum, or open-interest filter, and simple older tools can still be written as one complete procedure.
Entries in this reading3 entries

A state engine and a permission gate

The procedure is specified as a combination: a price moving average supplies the trend-state switch, and a volume condition is added so signals are not taken from price averages alone.

A moving-average-crossover is a close through a moving average of price used as the switch that declares or reverses trend state. Volume-confirmation is an entry filter that permits a new position only when session volume exceeds a recent average, treating elevated activity as participation rather than as a price target.

The trend-following premise

Trend-following is a directional procedure that assumes an established price trend is likely to persist and that stays aligned with that state until a later rule says the state has ended.

The premise treats a close above a moving average as a long state and a close below it as a short state, on the assumption that the force behind an established move tends to persist. Two assumptions are made explicit: underlying market forces tend to trend, and a moving average both marks when a trend is established and recognizes when that trend has ended.

The always-in-market baseline

Always-in-market is a ruleset in which a close through the average both exits the current side and immediately opens the opposite side, so the system never stands aside.

The baseline implementation entered long after a cross and close above a 20-session average, reversed short after a close below that average, and therefore remained continuously long or short.

Why a volume gate was added

Used in isolation, the 20-session close-through-average rule was judged unable to separate a minor fluctuation from the start of a new trend.

Volume above its recent average was treated as evidence that new information had reached participants and that the associated price move was more likely the beginning of a trend.

The abstention version

The combined entry required a close through the 20-session price average and that session’s volume above the average volume of the prior 10 sessions. Exits remained a close back through the price average alone.

After the volume filter was added, the system was no longer always in the market: a new position opened only when the crossover and the volume threshold occurred together. Abstention is the idle state created when the price average is crossed without volume confirmation, so no new position is opened.

Two design lessons

The write-up framed two design lessons: a trend-following engine can be paired with a volume, momentum, or open-interest filter, and simple older tools can still be written as one complete procedure.

Cumulative equity of the volume-gated T-bond strategy

The abstention version—20-session close-through-average plus a 10-session volume gate—compounds to about $35,472 over the 1988–1995 T-bond test, well above the always-in baseline. Points were read off the published cumulative-equity plot; the endpoint is the figure’s labeled $35,472.
The abstention version—20-session close-through-average plus a 10-session volume gate—compounds to about $35,472 over the 1988–1995 T-bond test, well above the always-in baseline. Points were read off the published cumulative-equity plot; the endpoint is the figure’s labeled $35,472.US Treasury bond futures · daily · 1988-01-01T00:00:00.000Z to 1995-12-31T00:00:00.000Z

One continuous T-bond contract, no commission or slippage; x is sequential closed trades on the modified (not always-in) rules. Digitized from the printed equity mountain, so intermediate levels are approximate.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
14 of 40 in the Volume confirmation track
19961-8 pp.Next on Volume confirmationConstructing four-state range-volume barsA facilitation ratio is the bar's high-low range divided by that bar's volume, and it is read only against the prior bar.
All readings on this track · 40 readings
  1. 1988Constructing volume-confirmation overlays on OHLC spreadsheet charts
  2. 1989Commodity advance-decline from delivery months
  3. 1989The most-active list as a three-layer breadth lab
  4. 1989Constructing a yield-curve volume-breadth composite
  5. 1989A bond-futures case study in support, volume, and confirmation
  6. 1989Volume-scaled price boxes and volume cycles
  7. 1990Futures-signed on-balance volume construction
  8. 1990Self-relative volume boxes for news-free breakouts
  9. 1990Broadening swings, demand tests, and volume filters
  10. 1991A same-session pressure test of breadth and volume share
  11. 1991Tick extremes that confirm double tops and bottoms
  12. 1991Constructing auction fuel from volume and open interest
  13. 1994Constructing On-balance volume with smoothing and timeframe confirmation
  14. 1996A volume-gated moving-average trend combination
  15. 1996Constructing four-state range-volume bars
  16. 1997Failed trade review of a descending-triangle breakdown
  17. 1997Construct a head-and-shoulders before the neckline break
  18. 2004Volume confirmation is not optional for a head-and-shoulders reversal
  19. 2007Constructing a three-factor volume-price confirmation filter
  20. 2007Constructing a three-condition moving-average entry with a volume filter
  21. 2008Breakout rules that wait for volume and liquidity
  22. 2011Screen futures liquidity with open interest and volume
  23. 2011Filter executable futures with liquidity and open interest
  24. 2012Rank listed futures by liquidity before a forecast chooses the name
  25. 2012Filter futures liquidity using open interest and volume
  26. 2013Pre-trade futures liquidity as an execution filter
  27. 2014A futures liquidity screen from range, open interest and volume
  28. 2014Evaluating futures contract liquidity before execution
  29. 2014Constructing defended price lines from volume clusters
  30. 2014Filter futures by equal-dollar size, open interest, and volume
  31. 2015Filter futures ideas by ranked contract liquidity
  32. 2017Screen listed futures for execution liquidity first
  33. 2017Filter futures orders by liquidity, open interest, and volume
  34. 2018Using open interest and volume to rank futures liquidity
  35. 2018Score listed futures as an execution menu before the setup
  36. 2019Filtering futures orders with liquidity, open interest and volume
  37. 2019Futures liquidity as an execution filter
  38. 2020A futures liquidity board as a pre-trade execution filter
  39. 2020Sequenced volume and golden-cross breakout rules
  40. 2020Use a listed-futures liquidity filter before execution
All 54 readings tagged Volume confirmation
Also on Volume confirmation5 readings