|Trader William Delaware Gann (1878-1955) reportedly had trading successes that would be the envy of anyone. Perhaps his observations are a by-product of point and figure charting back when charting techniques were most easily accomplished with charting paper. Gann fans, one of the techniques employed by Gann, use a predetermined set of angles and can be easily seen when using charting paper. They are used on charts where one unit of price equals one unit of time. Like Gann, Fibonacci numbers are also predetermined but their use does not require special charting techniques. Gann himself seems to have favored the use of Fibonacci numbers to find retracement levels. |
|Progressive Ohio [PGR], (top chart Figure 1) is representative of the charts I happen to pick at random (including Microsoft, QQQ, and AOL). The distinguishing features of PGR are an intermediate high followed by a gap down and then a downtrend to a low. There is an uptrend followed by downtrend, a gap up and then a peak in January 2000. The trends (Figure 1-top chart: U1 vs U2 and D1 vs D2) are somewhat parallel at times but in general if there is a pattern it's difficult to see. |
|Figure 1: PGR (NYSE) price chart without Gann and Fibonacci annotations (upper chart), and PGR with Gann and Fibonacci annotations (lower chart).|
|Graphic provided by: MetaStock.|
|Graphic provided by: Data vendor: eSignal<.|
|Now examine the bottom chart for the same time period. I have attached three Gann fans (Figure 1-bottom chart: red and green lines) to highs and lows. The fans have an equal rise and run. My choice was 32 x 32, because of work done by J. M. Hurst - and will be more evident in Part II. Gann fans provide trend resistance and support. The red Gann fan attached to the peak on the left, July 1999, shows resistance and support to downtrends. The coincidence of the price downtrend with one of the Gann fan lines is not unique to PGR; it happens on a number of charts. I have labeled the matching downtrend and fan line as A.|
Gann fans use the angles of 82.5, 75, 71.25, 63.75, 45, 26.25, 18.75, 15, and 7.5 degrees using a grid that has equal rise and run. Each grid unit of the x-axis is equal to a unit of the y-axis. In order to see these angles, take a piece of graph paper and mark off a section that is eight units high by eight units long. Let the lower left hand corner be the origin. Drawing a line from the origin to the upper right hand corner makes a 45-degree angle with the horizontal.
Gann considered this the most important angle in terms of trend resistance and support. Forty-five degrees is alternatively referred to as the 1 x 1 and corresponds to the point that is eight units from the origin to the right (run) and eight units up (rise). Starting at the origin, 2 x 1 is eight squares to the right and four squares up (26.25 degrees), and 1 x 2 is eight squares up and four units to the right (63.75 degrees). Get the 3 x 1, and 1 x 3, lines by going just 1/3 of the way up or 1/3 of the way to the right. They (1 x 3 and 3 x 1) are the exception in drawing the lines from the origin to a grid coordinate along either the top or right edge of the eight by eight square.
The next event is that the downtrend remains in tact and when it reverses it does so with a gap up. By coincidence, a Gann fan line marks the break in downtrend resistance. I call that event B. Now the price action is in the adjacent fan wedge to the right of the one where I started. It is a broad fan wedge. There are not many options at this point to see how uptrend U1 (see upper chart for label) will progress and end. You can attach a fan at the low point and the prices do trend up along the fan lines.
By mid-September 1999 it's probably fair to say that PGR had hit a low at the beginning of September. Suppose at mid-September you were to attach Fibonacci retracement lines to the peak in July and the low in the beginning of September. The result is a series of horizontal lines showing levels of retracement starting with 0% at the peak in July and going to 100% at the low in the beginning of September. The significant levels of Fibonacci retracement are 23.6%, 38.2%, 50%, 61.8%, and 100%. Now notice that the point where price deviates from the Gann fan line coincides with the 61.8% Fibonacci retracement level. I label this as event A1.
Using Fibonacci time zones (corresponding to 1, 2, 3, 5, 8, 13, etc. trading days), which start at the PGR peak in July 1999, there is coincidentally a match with the September intermediate peak and the Fibonacci retracement at 61.8%. I call that event C. Now PGR retraces back down to the 100% Fibonacci retracement line. I call that event D.
As it turns out, PGR is going through a double bottom. As price rises from the October low it takes an enormous gap up and lands close to the intersection of the Fibonacci 61.8% level and one of the Gann fan lines. I label this as event E. Attaching a Gann fan at the second double bottom, again using 32 x 32 for rise and run, one of the Gann fan lines acts as resistance, event F, and then moves sideways to the adjacent fan line and continues on up, all of which I call event G. Price moves sideways using the Fibonacci retracement level of 23.6% as support, which I call event H (which also has two retracement lows at the intersection of two Gann lines). Finally, in February 1999 five events coincide: a Fibonacci time zone line, a Fibonacci 61.8% retracement level, a fan line from the peak in July, a fan line from the low in October, and a turning point in price, which I call event I.
Are events A, A1, B, C, D, E, F, G, H and I all just coincidences? You could say that with enough lines you could hit a barn. But I always use 32 x 32 or 16 x 16 and see similar results. I always attach the fan origins to an intermediate or major high or low. You could say that you don't know ahead of time which fan line to choose. No, but it's clear that a fan line can "accidentally" act as a resistance or support line for a price trend. Or is "accidentally" just a little too patronizing when it's a price trend that's being followed? Clearly there are lines that don't count, so does that mean you would expect an event for every line? In other words, if you argue that there are some lines that don't come into play, then you are essentially saying every line and intersection has to have a significant event; that seems extreme in the other direction. I think the compromise lies in realizing if Gann fan lines work as support and resistance trend lines then some fan lines will offer little support/resistance while others offer more.
You can't dismiss all the events as just coincidences because using the same rules give similar results. The issue is reliability. The fact that some events don't take place would make a trading system based on Gann and Fibonacci alone difficult to trade. So perhaps the answer is to use Fibonacci and Gann in conjunction with other indicators.
One factor worth mentioning is that one of the Fibonacci time zone lines (12/6/2000 line) for the Fibonacci time zone started with the July 1999 peak happens to nearly hit a peak (12/11/2000) for PGR in December 2000. Also a Fibonacci time line (4/4/2000) for a time zone started at the low in October 1999 exactly hits an intermediate high (4/4/2000). Given that Fibonacci time zone lines get more spread out as time moves out and therefore the number of lines gets less and less, the chance of coinciding with a significant event becomes less and less.
For Part II, I'll go through QQQ and show another option in attaching fans. Elliott waves anyone?
|Company:||Technical Analysis, Inc.|
|Address:||4757 California Ave SW|
|Seattle, WA 98116-4499|
|Phone # for sales:||206 938 0570|
|Fax:||206 938 1307|
Traders' Resource Links
|Charting the Stock Market: The Wyckoff Method -- Books|
|Working-Money.com -- Online Trading Services|
|Traders.com Advantage -- Online Trading Services|
|Technical Analysis of Stocks & Commodities -- Publications and Newsletters|
|Working Money, at Working-Money.com -- Publications and Newsletters|
|Traders.com Advantage -- Publications and Newsletters|
|Professional Traders Starter Kit -- Software|
Click here for more information about our publications!