Market Cycles and Fibonacci

The relevance of trading with TIME-cycles alone is far less accurate than forecasting with PRICE. But its relevance will increase if, as the forecasted times approach, price patterns and momentum indicators show signs of reversal. Also, when trading markets like futures and commodities, the capital required to cover risks can be large due to leverage. Therefore, these long-term cycles only provide useful information in a limited capacity.

I discuss cycles here to illustrate:

• How several long-term cycles have accurately predicted significant market turning points.
• The amazing mathematical "coincidences" I have discovered while searching for an accurate market-timing method.

You may wish to bare this information in mind if trading long-term markets, such as stocks, possibly options or perhaps long-term index futures where a turning point is possible?

In the final two parts of this course (19 and 20) I will go through many examples of trading, on different time frames, using all the information shown throughout the course: A kind of "How to put all the information together".

For now, I hope you find this cycles information as fascinating as I do.

Below are some cycles that have been found by scientists and astronomers when studying the universe. Numbers that we have seen through this course (such as numbers that are or are near to Fibonacci numbers, or Gann multiples and fractions, are highlighted).

Note a tremendous amount of Fibonacci relationships!

Also, one of Gann’s major techniques for market timing was to use fractions of a circle, specifically into quarters, eighths and thirds, to count the number of days/weeks/months between highs and lows. For example, the circle has 360 degrees, 90 is one quarter, 45 is one eighth. Important numbers to count between highs/lows are therefore 30, 45, 60, 90, 135, (90 + 45), 150, 180, 210, 225, 270, 315, 330 and 360.

Rounding up one eighth of 90 is 11, two-eighths is 22, three-eighths is 33, 45, 56, 67, 78 and 90. These are other numbers to look out for.

1.09 (Fibonacci 1. One plus 9/100)

1.5 (Gann 150)

2.7 (Fibonacci 2.618. Gann 3 x 90 / 100)

3.5 (Gann 350%, 360 degrees / 100, and 33 is 3/8 of 90. Fibonacci 34 / 10)

5.5 (Fibonacci 5 and 55. Gann 56 is 5/8 of 90)

8 (Fibonacci 8)

10.2 (Gann 10, and 90 plus one-eighth of 90 or eleven = 101, ten-times Fibonacci 1)

15 (Gann 1.5 x 10)

12 (Monthly cycles, 12x12 is Fibonacci 144)

17 (7days in a week + 10)

22.2 (Fibonacci 21, and 2.236 x 10. Gann 22 is 2/8 of 90)

26 (Fibonacci 2.618 x 10. 25 is a Gann ¼ of 100)

34 (Fibonacci 34. Gann 33 is 3/8 of 90)

45 (Gann 45 degrees, half of 90 degrees, close to half Fibonacci 89)

59 (near to Fibonacci 55, Gann 56 is 5/8 of 90)

85 (near to Fibonacci 89, Gann 90)

96 (near to Gann 100%, Fibonacci 8 x 12)

178 (180 degrees in a circle, Fibonacci 377 – 200 = 177)

200 (Gann 200%, Fibonacci 2 x 100)

400 (Gann 400%, Fibonacci 2 x 200, near to Fibonacci 38% of 1000)

600 (Gann 200% x 3, Fibonacci 3 x 200, near to Fibonacci 62% of 1000)

900 (Gann 300% x 3, Gann 90 degrees x 10, Fibonacci 89 x 10, or 89% of 1000)

Humanistic, Historical Points (in years):

4.3 (Fibonacci 4.236)

5.2 (Fibonacci 5, and 55 dived by 10)

7.1 (7 days in a week)

10.5 (Gann 10 multiple, Fibonacci 5 x 2)

12.5 (Fibonacci 13, Gann 12)

16.1 (Fibonacci 1.618 x 10)

22.0 (Fibonacci 21, Gann 2/8 of 90)

35 (Fibonacci 34, Gann 350%)

55 (Fibonacci 55, 56 is 5/8 of 90)

130 (Fibonacci 13 x 10)

170 (1.62 x 100, 168 is 90 + 7/8 of 90)

200 (Gann 200%)

263 (Fibonacci 2.618 x 100)

317 (Gann 350 - Fibonacci 34 = 316)

350 (Gann 350%, ten times Fibonacci 34 = 340)

Problems with Counting Cycles

As well as regular cycles there are random fluctuations in things, too. The random occurrences can camouflage the regular cycles and also generate what appear to be new, smaller cycles, which they may not be. If you are zealous enough you can find regularity in almost anything, including random numbers where you know that the regularity has no significance and know it will not continue. This is the problem with market-timing signals.

Also, many things act as if they are influenced simultaneously by several different rhythmic forces, the composite effect of which is not regular at all.

The cycles may have been present in the figures you have been studying merely by chance. The ups and downs you have noticed which come at more or less regular time intervals may have just happened to come that way. The regularity - the cycle - is there all right, but in such circumstances it has no significance.

The following examples illustrate this problem of cycles appearing/disappearing.

Cycles in the Stock Market

When forecasting stock market cycles, the cycles are influenced by random events. Cycles are inherently unreliable and their predictive value provides only specific probabilities when the suggested time period is approached.

Fixed time cycles are apparent in stock market tops and bottoms. But eventually a cycle will cease to continue. For example, the four-year cycle in the US stock market held true from 1954 to 1982 producing accurate forecasts of 8 market bottoms. Had an investor recognised the cycle in 1962, he could have amassed a fortune over the next 20-years. But in 1986, the cycle’s prediction of a low failed to provide a bear market and in 1987 its rising portion failed to prevent the largest crash since 1929.

Another cycle that may have disappeared is the 3-year cycle that began in 1975, forecasting lows in 1978, 1981, 1984, 1987 and 1990 – there was no significant bottom in 1993, 1996 or 1999.

Other long-term cycles (such as Kondratieff and Benner/Fibonacci) as well as Elliott Wave counts, suggest that the ultra long-term bull market may be coming to an end.

Therefore, many old or existing cycles may come to an end and new ones begin.

It is difficult at the best of times to recognise a cycle taking place, but with the high probability of a reversal in ultra long-term trend at the time of writing (March 2002) it is even harder to confirm. Nevertheless, the text below explains how a lot of cycles have been seen on the long-term US stock market.

Fibonacci Relationships in the Stock Market Cycles

1 year is a little less than 13 months, a little less than 55 weeks and a little less than 377 days. Thus a Fibonacci time period in one natural duration is close to a Fibonacci duration in another.

The Kondratieff Cycle is a common, often-quoted cycle of financial and economic behaviour that lasts approximately 54 years. This 54-year cycle is very close to a Fibonacci 55 number!

The 54 (55) year cycle was recognised by the Maya tribes of ancient Central America, the ancient Israelites, and rediscovered in the 1920’s by Russian economist Nikolai Kondratieff (hence the name of the Kondratieff Cycle.)

Dividing the Kondratieff Cycle of 54 years by 2 equals 27 years, and dividing by 2 again equals 13.5 years. This is near to a Fibonacci number 13 and, 13.5 years multiplied by 12 months equals 162 months – a Fibonacci 1.62!

Dividing 54 by 3 equals 18 years and dividing this by 2 equals 9 years, or 108 months. Dividing by 2 again leaves a smaller cycle of 4.5 years, which is 54 months – almost a Fibonacci 55!

Two-thirds of 54 equals 36 years. 5-times 36 years gives 180 years. This is the same as 180 degrees is half a circle, or half a planetary orbit.

All these periods are inter-linked by Fibonacci! How bizarre!

Remember that the proportion of two-thirds was used greatly by Gann. It is also near to a Fibonacci 0.618 ratio.

Let us take a look at a long-term chart illustrating the Kondratieff cycle:

Also, on the US stock market, the Kondratieff Cycle appears to subdivide into harmonic sub-cycles of between 16 and 20 years. The last set of sub-cycles saw US stock market lows in 1842, (+17) 1859, (+18) 1877, (+19) 1896, (+18) 1914, (+18) 1932, (+17) 1949, (+17) 1966 and (+16) 1982.

The diagram below is based on Samuel Benner’s cyclic discoveries but I have modified and updated it to fit the behaviour of the stock market.

It uses 3 cyclic periods to project each reversal point.

The first cycle goes: 8-years, 9-years, 10-years, and begins in 1902. The projected lows were forecast on 1902, (+8) 1910, (+9) 1919, (+10) 1929, (+8) 1937, (+9) 1946, etc.

The next cyclic periods project reversals in years of 16-years, 18-years, 20-years (i.e. double the period of the first cycle). Starting with an 18-year period from 1903, this cycle forecast lows in 1903, (+18) 1921, (+20) 1941, (+16) 1957, (+18) 1975, etc.

The next cyclic period again uses the 16-18-20-year counts, but begin in 1913. This cycle projected market turning points on 1913, (+20) 1933, (+16) 1949, (+18) 1967, (+20) 1987, etc.

If you compare this Benner Cycle with a long-term stock market chart, you will see how it predicted many of the historic high and low turning points (Figure 335.)

THE REST OF THIS ARTICLE IS ONLY AVAILABLE TO THOSE WHO PURCHASE OUR TRADING COURSE