Skip to main content
Track Double top and bottom
20 / 32
Search the archiveLibrary
Library

2004issue C061-4

A case study of the shark-attack Fibonacci retracement

This archive article restates a historical workflow that plots a Fibonacci retracement, waits on a double-top or double-bottom test, and places a stop-loss at the second extreme. An editorial reading treats a false breakdown or breakout as a two-part hypothesis, and names the 1.272 to 1.618 extension a reversal zone only when that stop can be placed.

  • A Fibonacci retracement is plotted from an identified swing high and swing low at 38.2%, 50%, and 61.8%, with 25% and 75% used only if those first levels do not hold.
  • A long shark-attack starts at a downswing low, retraces, then either holds as a double bottom or extends near 1.272 and 1.618 of the prior correction before a sharp rally.
  • The historical long entry is a break of the previous bar's high, with the second lower low used as the stop-loss.
  • Editorial reading: name the 1.272 to 1.618 extension a reversal zone only after a double-top or double-bottom test, and only if a stop-loss can sit at the second extreme.
Entries in this reading3 entries
Current trackDouble top and bottom68 readingsFibonacci retracement97 readingsStop-loss order144 readings

How a Fibonacci retracement is built

A Fibonacci retracement starts from an identified swing high and swing low. The common levels are 38.2%, 50%, and 61.8% of that completed swing. If those levels do not hold, 25% and 75% are available.

The 0.618 ratio is obtained from successive Fibonacci numbers such as 21/34, 34/55, and 55/89. The 0.382 level is 1 minus 0.618. The 0.5 level is the mean of 0.382 and 0.618. That 0.618 / 1.618 relationship is the golden-ratio basis for the common retracement and extension levels.

Trend depth and confluence

In strong bull or bear markets the archive describes prices often retracing no more than 25% to 38.2%. In moderate trends they can retrace up to 50% to 61.8%. A break of the 61.8% to 75% zone is treated as a threat to the main trend.

The most significant support is described as confluence of two Fibonacci levels measured from two different swing lows to the same swing high. The most significant resistance is the converse, measured from two different swing highs to the same swing low.

The long shark-attack sequence

A long shark-attack setup assumes a downswing low at point 1, an upward retrace to point 2, then a test of the prior bottom. That test either forms a double bottom or breaks to point 3 near 1.272 and 1.618 of the prior correction before a sharp rally.

In this archive usage, a shark-attack is a failed continuation after a break of a prior high or low. Price extends about 1.272 to 1.618 of the last correction, then reverses sharply as late breakout traders are forced out. A double-top or double-bottom test is the retest that either holds and reverses or fails and continues, and it is the setup that can become that reversal.

Entry, stop-loss, and crowded continuation

The long entry is described as occurring when the high of the previous bar is broken. The second lower low is used as the stop-loss. On a short, the matching stop-loss sits at the second higher high, so the reversal idea is abandoned if that extreme is taken out.

The archive warns that simple breakout and breakdown trades should be treated with caution whenever a crowded consensus expects continuation, because a shark-attack reversal can follow that consensus.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
20 of 32 in the Double top and bottom track
20041-5 pp.Next on Double top and bottomConfirming index turns with envelopes, divergence, and breadthUse the daily moving-average envelope only to mark whether an index swing is at an extreme relative to a typical range around the 21-day average.
All readings on this track · 32 readings
  1. 1988Reaction length as a trend integrity test
  2. 1991Sold-out double bottoms as a three-gate inventory test
  3. 1991A breadth classifier for V-bottoms and W-bottoms
  4. 1991Precomputed price-ratio clusters and double-top tests
  5. 1992Bond turning points as a regime check on equity double tops and breakouts
  6. 1992Commodity-bond ratio as an equity regime overlay
  7. 1992Gold lead confirmation for commodity-index turns
  8. 1994Constructing the thousand-line advance-decline indicator
  9. 1995Evaluating zero-line patterns on a breadth-price oscillator
  10. 1996Constructing double tops from a resistance retest to a trough break
  11. 1996Four-stage double-bottom construction
  12. 1998Double-bottom confirmation and stop placement
  13. 2000Two-bar reversal construction
  14. 2001Constructing double tops from failed resistance retests
  15. 2002Eve-Eve double bottoms: width, confirmation, and overhead resistance
  16. 2002Constructing Eve-and-Eve and classic double bottoms
  17. 2003Eve-Adam double bottoms as a two-step classroom test
  18. 2003Shape contrast then breakout confirmation in Adam and Eve double bottoms
  19. 2003Reading cyclical bottoms inside secular bear regimes
  20. 2004A case study of the shark-attack Fibonacci retracement
  21. 2004Confirming index turns with envelopes, divergence, and breadth
  22. 2005A five-wave euro/dollar case and the support that still had to fail
  23. 2007Constructing commodity seasonal indexes for regime context
  24. 2009Constructing rounded and double-top short setups
  25. 2010Hourly pattern entries, exits, and abstention as one playbook
  26. 2016Ugly double bottom after a yearly low
  27. 2016An unconfirmed stock double bottom next to a confirmed index
  28. 2016Constructing a range-midpoint moving average
  29. 2017Evaluating whole-dollar delays on pattern breakouts
  30. 2018Volume-confirmed bottoms and breakouts with moving averages
  31. 2018Evaluating double bottoms with a locked stochastic confirmation
  32. 2019Forex pairs as relative value: yield spreads, support, and a double bottom
By fieldWaves and ratios90 readingsBy year2004107 readings
All 68 readings tagged Double top and bottom
Also on Double top and bottom5 readings
1991Precomputed price-ratio clusters and double-top tests1998Double-bottom confirmation and stop placement1999Dead-cat bounce, rollover, and failed reversals2016Ugly double bottom after a yearly low2019Structure invalidation before comfort-stops