The Fibonacci method is one tool in a group of methodologies (such as Elliott Wave and Gann) that seeks to tap into and exploit an underlying universal structure of all markets. Simply put, if this structure provides the apparatus within which markets move, then measuring it and the market’s place within it will give you an exceptional edge.
While Fibonacci is a good method for uncovering the market’s hidden order, and reliable in the right hands, it must be analyzed in the proper context. Traders also should maintain realistic expectations of what Fibonacci can, and cannot, do.
Fibonacci numbers are derived from the work of Leonardo de Pisa de Fibonacci, who was a 13th century mathematician. He popularized a series of numbers, later named after him, that increases by the sum of the preceding two numbers. It is, 1 (0+1), 2 (1+1), 3 (2+1), 5 (3+2), 8 (5+3) and so on until we get 21, 34, 55, 89, 144, 233, etc.
We know that markets don’t go straight up or down, making them difficult to predict. Our interest is in universal price movement, the understanding of which gives the trader a better chance to manage risk. But the Fibonacci series alone does not help manage risk in terms of price retracements. However, it does correspond to natural, universal price movement in terms of time.
According to Fibonacci methodology, a move from point A to point B in any time frame can retrace either 38.2%, 50%, 61.8% or 78.6%. Other important numbers also come into play here. Briefly, the golden spiral number of 0.618 squared is 0.3819 and the square root of 0.786 is 0.886. That’s why those numbers work as retracements. While 50, or one-half, does not have a tie to Fibonacci, W.D. Gann, another analyst who pioneered research in the order of the markets, popularized the hypothesis that a move will correct half of its progress before it rejoins its main trend.
Of course, theory means nothing without application. There are ways we can put Fibonacci’s lessons to work for us.
Perhaps first and foremost is identifying Fibonacci retracements. These work best as an intermediate- to long-term forecasting tool. One of the textbook examples you will see is in the SPX covering the period from the bottom of the bear market in 2009 through the low in October 2011 (see “Back for more,” below). The first phase of the bull peaked in April 2010 and retraced approximately 38.2% of the move, which bottomed in July 2010. The bull resumed to May 2011 where it pulled back throughout the summer and found a trading low in October 2011, very close to the 38% retracement of the larger move once again.
An important point to consider is that while many traders are looking for perfection and precision, the reality is that many times the retracement will be close to the ideal but overshoot or undershoot the target by a modest amount. Also, the Fibonacci retracement analysis will be more reliable if it lines up with another universal methodology, such as Andrews’ pitchfork channels (see “Andrews’ pitchforks and the price failure rule,” June 2011).
The market doesn’t ring a bell at highs or lows, but does provide a hint in terms of various clues. The degree of the retracement is the market’s way of playing its hand. A 38% retracement is considered a minimum correction and in an unscientific study (observation and not statistical), it has been evident dating back to the Great Depression. In instances where the Dow pulls back 38% on a weekly time frame, odds are enhanced greatly that the market at least will touch the prior high. No other retracement level comes close in reliability.
A 61% retracement portends overall weakness to the main trend, but it has value as a trading leg. In fact, many traders use the Gartley trading method, devised by H.M. Gartley, which looks for an A-B-C pattern that places stops going at the 61% retracement. After a 61% retracement of the main trend, it is not a requirement that the pattern make a new high beyond the old low for the trader to be consistently profitable using that method.
However, in terms of forecasting, if an important index or sector does play its hand, the trader can leverage that information and gain a different type, but still excellent, of edge. The banking index BKX gives us such an example (see “Banking on a bounce,” below). It also shows us that a retracement does not need perfect precision to work.
The BKX low in 2009 was 17.76 with a peak at 58.81 in April 2010. The range is 41.05 and 0.618 of the range is 25.36. Subtract 25.36 from the 58.81 high, and we get 33.45. The low on Oct. 4, 2011, was 32.56, which is less than a full point off. As it turned out, that low was also in the 161 day trading window to its February 2011 peak.
If a retracement can cluster with a Fibonacci time window, it has a much greater chance of working out. Time is an important piece of the puzzle, and markets will respond to Fibonacci (and other) time windows with regard to both trading and calendar days. In terms of leverage and banks, if an important sector turns, chances are the overall market does as well.
Fibonacci retracements also can work in highly charged emotional markets. An example materialized at the emotional turn in August when the Nasdaq futures bottomed. The price action bottomed and rallied 159 points in about 15 hours. It then dropped approximately 99 points in roughly two hours. How does a trader navigate this kind of volatility? A close look at the action shows the retracement leg stopped going down at exactly 61% of the first move off the bottom. There is no other technical explanation of why the pattern turned where it did.
Because of hits like these, many traders will rely on the Fibonacci retracement to target intraday turns. While the trader will get some successes overall using retracements as intraday trading targets, the analysis has more misses than hits. For targeting intraday turns on charts such as E-mini stock indexes or forex, Fibonacci retracements are not as effective. Retracements tend to work better on larger time frames.
As we’ve seen, the Fibonacci numbers aren’t used as retracements; it’s the square root relationships that work as retracements. The best use of pure Fibonacci numbers in financial markets is in market timing. Markets routinely change direction in all degrees of trend in time windows, measured by either Fibonacci numbers or golden spiral derivative relationships. For instance, the same low in the NDX, which was an important market turn in August, missed a perfect Fibonacci 610 trading days by a day-and-a-half when counting partial bars.
One of the most important Fibonacci timing examples from 2011 was in the Japanese earthquake/tsunami/nuclear disaster. Who can forget the scenes and emotions of those days? The Nikkei literally crashed, and when combined with market sentiment, it felt like there would be no bottom. As it turned out, the Nikkei moved similarly with the BKX index, which topped in April 2010 and never made a new high in 2011 like many markets did.
The silver lining to the cloud was that the Nikkei responded to the Fibonacci 233 trading-day window (see “Trading on edge,” below). This was significant because anyone who realized this timing relationship would have at least understood that when combined with sentiment, the underlying pattern was ripe for a turn. For world markets, Japan was driving the bus in this particular sequence. Not only was the Nikkei in free fall, but the dollar vs. yen was at a multi-decade low. Anyone looking for a trading edge could have found one simply by understanding the cycle work on the Nikkei.
Finally, like its retracement counterpart, the Fibonacci time cycle window is no panacea. Markets can elect to change direction in these windows or easily bypass them. The trader who elects to front-run the bars and place an order just because he sees a window approaching (in any time frame) is likely to be disappointed.
It’s important to combine the Fibonacci turn window with another method, such as key support or resistance areas or an extreme in market sentiment, to have success. This also is true for the retracement. You must validate the retracement or the time window by effective use of candlestick methodology. A known formation such as a candlestick morning/evening star or engulfing pattern is needed to validate any turn.
The challenge that plagues many beginning-to-intermediate-level traders is they think all they need to do is find the retracement or time window to achieve a high degree of success. The truth is that while Fibonacci is an excellent methodology, it is best used as a tool, not a trading system. The Nikkei example likely worked because sentiment was an extreme.
On the flip side, markets will come to extremes and become even more extreme in sentiment as they move toward the time window. The best turns happen as a result of several factors, not any particular one. As there is no such thing as the Holy Grail, universal methods such as Fibonacci are best suited to be supporting data in a larger context of several factors.
Jeff Greenblatt is the author of “Breakthrough Strategies For Predicting Any Market,” editor of the Fibonacci Forecaster, director of Lucas Wave International and a private trader.