Skip to main content
Track Adaptive moving average
8 / 24
Library

1998issue C011-6

Testing price-channel breakouts with a lag-aware adaptive average

A sideways market is mapped as a price-channel so its edges and breakouts can be tested. An adaptive moving average is kept as a lag-aware trend filter around that structure, not as a standalone forecast.

  • Sideways action is treated as a price-channel because a conventional moving average weaves through closes and produces frequent false turns.
  • The adaptive-moving-average is used only as a lag-aware filter: paintbar-state marks bull, bear, or pause closes, and a horizontal VIDYA 21,5 is what defines the channel.
  • A valid upside breakout must rise through the 100% channel-level, hold a short consolidation above it, and then confirm with later closes, including an equal close at VIDYA 21,5.
  • The one-close-rule keeps a bull trend intact after a single close through VIDYA 21,5, and an upside move through mapped channel-levels without the required pattern is treated as a likely bull trap.
Entries in this reading3 entries

A sideways market as a testable channel

Editorial note: TradersWeek presents this archive workflow as a way to treat a sideways market as a defined price-channel whose edges and breakouts can be tested, then to use an adaptive moving average only as a lag-aware trend filter rather than as a standalone forecast.

The archive method sorts market direction as up, down, or sideways. Sideways action is flagged as a price-channel, a stretch in which the adaptive average flattens and the market trades between mapped percentage levels instead of advancing in one direction. A conventional moving average is a poor tool for that state, because the average weaves through closes and produces frequent false turns.

How the adaptive moving average changes lag

The adaptive-moving-average in the archive is VIDYA, a variable-length average that shortens or lengthens its effective lag as directional momentum rises or fades. VIDYA is constructed from a momentum oscillator that sums a lookback of up and down periods, takes the absolute value of that ratio as an index, and uses the index to adjust the smoothing of an exponential average. The construction lets the line lengthen lag when directional momentum is strong and shorten lag when momentum collapses.

Length, smoothing, and a day-trading pair

Shorter length and smooth settings track nearer-term price more closely than longer settings. A 55-period VIDYA that smooths the last 21 closes is described as less specific than a 13-period VIDYA that smooths the last eight closes. The day-trading pair used on five- and eight-tick S&P 500 charts is VIDYA 21,5 for the broader trend and VIDYA 8,5 for the immediate trend.

On a December S&P 500 comparison, VIDYA 21,5 stays closer to current prices and turns more promptly than a simple 21-period average, which tracks at a wider distance and changes direction more slowly.

Paintbar-state as a trend filter

The archive uses a paintbar-state, a three-color close-to-band reading that marks bull, bear, or pause conditions without drawing extra bands on the chart. A close above the upper VIDYA band is painted as a bull trend, a close below the lower band as a bear trend, and a close between the bands as a pause or possible reversal. Yellow pause bars are treated as a cue to tighten stops and use caution rather than to exit the position.

When a price-channel is defined

A price-channel is defined when VIDYA 21,5 turns horizontal instead of rising or falling at an acute angle. After that flattening, channel limits can be mapped as channel-levels: fixed percentage stations drawn from the channel range and used as support, resistance, and breakout checkpoints. Later breakouts are judged against those limits.

What counts as a valid breakout

In this workflow, a breakout is a move through a mapped channel extreme that is accepted only when a short consolidation holds beyond that extreme and later closes confirm the new direction. A valid upside breakout pattern requires price to rise through the 100% channel-level, a short consolidation to hold above that level, and later closes to confirm the new trend. Those confirming closes include an equal-close-rule check: a close at VIDYA 21,5 is treated as evidence that the uptrend remains intact.

Keeping a trend intact

A single close through VIDYA 21,5 does not, by itself, end a bull trend. The one-close-rule is a confirmation check that refuses to call a trend ended after a single close through the adaptive average. The trend stays intact until additional structure, such as a sharp sideways turn in VIDYA 21,5 plus neutral bars, argues that a new price-channel may be starting and the long can be exited.

When the breakout pattern is missing

Absence of the required breakout pattern is itself useful. An upside move through the mapped channel-levels without the consolidation and confirmation conditions is treated as a likely bull trap rather than a confirmed breakout.

Editorial reading: the falsifiable object is the mapped price-channel and the breakout pattern at its extremes. VIDYA is the filter that shows when the market is trending, pausing, or flat enough to draw channel-levels.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
8 of 24 in the Adaptive moving average track
19981-3 pp.Next on Adaptive moving averageConstructing filters by nesting offsets and variable weightsAnchored momentum is 100 times a ratio-minus-one that compares the close, or an exponential average of the close, with a simple moving average used as a fixed reference window.
All readings on this track · 24 readings
  1. 1991Building variable-length moving averages from partitioned price changes
  2. 1991Variable-length moving average from change dispersion
  3. 1992Constructing volatility-adaptive exponential smoothing
  4. 1995Constructing an adaptive moving average with an efficiency ratio and filter
  5. 1995Two-gate breakout confirmation with adaptive averages
  6. 1995Building momentum-scaled adaptive moving averages
  7. 1995Adaptive length as a construction choice inside exponential smoothing
  8. 1998Testing price-channel breakouts with a lag-aware adaptive average
  9. 1998Constructing filters by nesting offsets and variable weights
  10. 1998Constructing an efficiency ratio adaptive average and entry filter
  11. 2001Encoding candle structure as a numeric filter
  12. 2001Adaptive averages driven by cycle-phase speed
  13. 2005Constructing an adaptive moving average from a fractal-dimension weight
  14. 2005Range-dimension adaptive exponential filter
  15. 2010Constructing simple, exponential, and adaptive averages
  16. 2010How a price-hugging smoother is assembled from ordinary averages
  17. 2013Evaluating an adaptive moving average against a same-window moving average
  18. 2016Three-layer confirmation: adaptive average, stochastic relative strength index, and stop-and-reverse
  19. 2017One-alpha reverse-path exponential smoothing
  20. 2018Pair two adaptive averages to filter swing turns
  21. 2018Two Adaptive moving averages as a confirmation pair
  22. 2018Constructing an adaptive filter for adoption-cycle reversals
  23. 2018Constructing a deviation-scaled adaptive moving average
  24. 2020Walk-forward and adaptive averages as two tests of the same trend
All 28 readings tagged Adaptive moving average
Also on Adaptive moving average5 readings