CISCO


CISCO Futures
1-303-306-1521 1-800 800 7227 Fax 1-303-306-1572
http://www.cisco-futures.com
dljones@cisco-futures.com



Candlesticks, a Critique
By Donald L. Jones©

There is no 'trading university'. A potential trader typically looks for training and educational courses on the internet or through a brokerage firm. Most of these sources have a goal, brokerages to get the new person trading as soon as possible; purveyors of courses to get the newbie to take their course. Neither route seems to be very successful, since it is reported that some 90 to 95 percent of new traders fail within their first year.

The courses often teach a goodly amount of worthwhile information; a respect for leverage, a reasonable approach to risk management and so forth. Most courses fail where the rubber meets the road, at the trading model level. Most analysis schemes rely on some sort of smoothing, e.g. moving averages and oscillators. Some use an elegant display method called 'Japanese Candlesticks'.

Candlesticks Defined
Darrell Jobman, the former editor of Futures Magazine (ref1), discusses Candlestick charting, pointing out that they "emphasize the open and close and put less weight on the high and low". Further, compared to standard bar charts, "the open is much more significant for the candlestick chartist". Also, "the difference between the open and close makes up the 'body' of a time period's trading range. "It actually often looks like a candle with a wick or a tail known as a 'shadow' at either end. "If the closing price is above the opening price, the body is clear or white:, if the price at the close is below the open, the body is solid, or black." Jobman goes on to say: "Generally a series of clear or white candles indicates an uptrend or bullish market...." And he states: "A series of solid or black candles indicates a downtrend or bearish market....." Lastly, "The color and length of the candlesticks themselves provide a quick visual clue as to whether buying or selling dominates the market."

Jobman continues: "...any candlestick analyst will tell you one candle is not enough to make a trading decision." Further, "(candles) should not be viewed in isolation apart from other technical evidence".

Finally, Jobman lists various candlestick patterns and what they suggest: note, this commentary is much more abbreviated than Jobman's explanation.
1. Doji. Open and close same, unsure market
2. Spinning Tops. Small real body, somewhat like Doji
3. Hammer. Also tight open/closing range. Suggests a bullish turn
4. Hanging Man. Like Hammer, except comes at top of a market trend
5., 6., 7., 8. Stars. Small real body leaves a gap (4 kinds of stars), market is ready for reversal
9. Dark Cloud Cover. Bearish damper on an uptrend
10. Piercing Line. Bullish converse of Dark Cloud Cover
11., 12., 13. Bearish Engulfing Pattern (3). Body is larger than previous body
14. Harami. An inside day, suggests change may be imminent
15. Harami Cross. Harami including a Doji, suggests losing momentum

Although there are many candlestick patterns, there is a certain firmness to the definitions. The forms defined are real and, to some degree, measurable. It seems possible to develop a computer code of a series of open, high, low and closes, tracking a market from a candlestick viewpoint. After all, a candlestick itself is not very complicated, as the example below illustrates. 



                         |        high    471:0
                         |
                       -----      close   468:6
                       |   |
                       |   |
                       |   |
                       -----      open    465:4
                         |
                         |
                         |        low     465:0


Giovanni Maiani developed such a program, reporting his results in Stocks & Commodities Magazine in November 2002. He reports analyzing 575 American stocks over a period of 15 - 20 years. He found more than 6.14 million candlestick formations. In his words:
"The most reliable, long lower shadow, is present 1.9% of time, and anticipates a declining session 49.12% of time; a rising session 40.61% of time. For traders who are quantitatively based, candlestick patterns are not terribly useful"

Subsequently, Robert Colby, writing in SFO Magazine, Aug., 1996, found "No Major edge for Candlesticks". Further, he states: "Nevertheless, the candlestick patterns tested here failed to beat the simplest trend-following standard of comparison on a significant basis."

Steve Nison, the creator of Candlesticks, responded (S&C Jan 2003, p74) that Candlesticks are "A tool, not a system", and that you need to know the trend to use them.

However, in his article in Active Trader Magazine, January 2003, 'Candles Shed Light on the Market', one of the benefits of candle charts, Mr. Nison states, "Candle charts not only show the trend of the move, as does a bar chart,...."
http://www.candlecharts.com/about-press-room.html

We feel that if you already know the trend there is little reason to use something that we, at least, would use to find the trend. From Jobman's description, it would seem that, unlike Mr. Nison's letter to the editor, traders do look to candlesticks for directionality. As is clear from his article in Active Trader, Mr. Nison does claim candle charts show directionality.

Recognizing trends is important. The sooner, the better. After all, it is the trader's job to get aboard a directional move and to exit when congestion sets in. It certainly appears that traders think they are divining directionality from their candlestick charts, and the courses that use candlesticks for their modeling are doing the same. If Mr. Maiani's reserch is to be believed, candlestick charts are not a very good way to find directionality.

An Aside on Technical Analysis:
In the Japanese Candlesticks technical method, the basic element is a cylinder with open and close as cylinder limits; with the high and low spiking above and below the open/close base. It is well known and consistently observed in profiles that the high and low prices are rejected quickly by the market. So it appears that candlesticks are giving most weight to prices the market quickly rejects. Consequently, it is no surprise that Maiani found candlesticks to be of little use for trading decisions. Candlesticks and other charts do have the value of offering a panoramic view of market history. While of little use for current, day-to-day trading decisions, a general view of one's market is worthwhile for understanding a market in general.
Market/Meta Profiles Background

Using the high and low prices for trading decisions, or emphasis in charting is simply making important decisions from poor data. Many technical calculations use the same data in much the same way and suffer the same defect.

CISCO teaches market analysis and trader training. Nothing in the CISCO courses suggests that candlesticks or any of the other standard 'indicator' methodologies can help a trader to success. Trading is a sophisticated profession and like other professions, there is a learning curve. A set of patterns like the candlesticks would have to be proved to be reliable and testable to be a valid part of the learning process. The only objective tests so far lead us to vote against candlesticks as a market guide.
Short Course



References
1. The Handbook of Technical Analysis, Jobman, Probus, 1995, pg 91



Return to CISCO Futures Main