Long Legged Doji Candlestick Pattern

THE LONG-LEGGED DOJI (RICKSHAW MAN), THE GRAVESTONE DOJI, AND THE DRAGONFLY DOJI

As shown in Exhibits 6.2 to 6.4, some doji have nicknames depending on if the open/close (i.e., the horizontal component of the doji) is at the low or high of the session or if there are unusually long upper and lower shadows on the doji.

A candle line with long upper and lower shadows and a small real body is called a high-wave candle.1 If such a candle line is a doji instead of a small real body, then it is called a long-legged doji (see Exhibit 6.2). It also has the nickname rickshaw man.

The doji portion of the long-legged doji shows the market is at a transition point. The long upper shadow shows the market had rallied during the session, but by session's end had backed off from these highs. The extended lower shadow visually depicts a market that had sold off during the session; by session's end it had been able, by the close, to recoup some of its lost ground. In other words, the market rallies, sells off, rallies, etc. It is a confused market. These long shadows hint, as the Japanese say, "The market has lost its sense of direction." As such, a long-legged doji is an indication of a market separating from its trend.

Gravestone doji

The gravestone doji (see Exhibit 6.3) is another distinctive doji. It develops when the open, low, and close are at the low of the day. This line is an example of how visually intuitive candles are. Even if you never saw an explanation of the gravestone doji, just by looking at it, would you think it was a bullish or bearish signal? Of course, the answer is bearish. With its extended upper shadow and close at the low of the session, we can graphically see that at some time during the session the stock had rallied and by session's end the bears dragged this stock down to the lows at the close. It can be viewed as the ultimate shooting star. The shooting star has a long upper shadow and small real body. If the shooting star's real body becomes a doji, then we have a gravestone doji.

The gravestone's forte is in calling tops. The shape of the gravestone doji makes its name appropriate. As we have discussed, many of the Japanese technical terms are based on military analogies. In Japanese candlestick literature it is said that the gravestone doji represents the gravestone of the bulls that have died defending their territory.

Dragonfly doji

The dragonfly doji is the bullish counterpart of the gravestone doji. The dragonfly has the open/close at the highs of the session. This means the market had touched much lower lows during the session, but had impressively managed to close at, or very close to, the highs. This is like the hammer, but the hammer has a small real body where the dragonfly doji has no real body since it is a doji.

In Exhibit 6.10 a long-legged doji appeared on October 23. By using the highest high between the doji and the prior white candle, we immediately obtain resistance near $88. The index

Exhibit 6.10. NASDAQ-100 Trust-Daily (Long-Legged Doji)

descended from this long-legged doji until the hammer three days later. In early November the market was ascending, but the two long-legged doji put a damper on any bullish outlook. A further sign of trouble emerged with the November 6 and 7 bearish engulfing pattern. Note how in the early part of October the stock was falling with doji during this descent. As previously discussed, doji, based on my experience, work better at calling tops reversals than bottom reversals.

An example of a gravestone doji is shown in Exhibit 6.11. Since the doji's high was the same as the prior session, it is also a tweezers top. As ominous as "gravestone" sounds, please don't make the mistake of thinking it implies a large move lower. While the gravestone doji, because of its extended upper shadow and close at the session low, does increase the chances for a turn, it doesn't forecast the extent of a potential decline. Candle signals are unexcelled for spotting the early reversals, not to predict the extent of a move. Don't make more out of candles than they are designed to provide. As the Japanese adage goes, "Don't climb a tree to catch a fish."

Exhibit 6.11. Inktomi-15 Minutes (Gravestone Doji)

Although that doji was a reason for caution, I would not have turned outright bearish on that gravestone doji. Why? Notice how the close of that doji was still over the late March 22 resistance area near $224. It would take the close on the session after the gravestone to turn the trend south. That session, with its close back under $224, proved what the doji had hinted at— that the new highs couldn't hold.

In Exhibit 6.12 some long-legged doji in mid April signaled that the prior downtrend was losing momentum near $6.75. The rally in early May made a rising window. But there was a problem on the day of the rising window on May 9: It was another long-legged doji. This signified that the bulls were not in complete control. The window's support was broken the day after this long-legged doji. This confirmed the potential for another down draft. When the stock got near the aforementioned $6.75 support area from mid April on May 23, it successfully defended it with a hammer. The rally from this hammer made two windows at 1 and 2. The stock corrected from the bearish engulfing pattern at B. The decline from this bearish engulfing pattern found stabilization near $8.75 to $8.25 with a series of dragonfly doji. Since the third dragonfly didn't have the open/close at the high, but near it, I view it as a variation of the dragonfly doji. These dragonfly doji confirmed support at the windows at 1 and 2. Final confirmation of the support came with the June 28 hammer.

Exhibit 6.12. 3Com-Daily (Dragonfly Doji)

The general concept with doji (and for all candle signals) is that one should look at what happened before the signal. For instance, a doji in a rally is a potential reversal. Therefore, there needs to be a rally to reverse. This means that doji have little forecasting implications if they are in a trading range environment since there is no trend to reverse. The Japanese aptly call a market that is locked within a range a "box."

The doji in Exhibit 6.13 is reflecting on a micro scale what the trading range environment on a more macro aspect is showing: The market is undecided. With no trend to change, the doji in Exhibit 6.13 has no forecasting implications, except for the fact that it is confirming a trendless environment. An exception to this is if a doji, while still in a trading range, is at the top or bottom end of a range. As such, it is confirming resistance or support and could be a useful signal.

Exhibit 6.13. Doji in a Box Range

In Exhibit 6.14 let's look at three doji in relation to their preceding trend. Doji 1 is in the middle of a box range. Consequently, this doji doesn't have any forecasting implications since there's no trend to reverse. The same scenario is true with the dragonfly doji at 2. The doji at 3 is very different because of where it appears. This doji comes after a rally that placed the stock into an overbought condition. As such it has reversal implications. The two doji following doji 3 are echoing what the first doji at 3 told us—that the stock was running out of bullish force. To recap: Where the doji is in relation to its preceding trend is of pivotal importance.

Exhibit 6.15 illustrates a point alluded to previously in this chapter: Doji during descents (what I call Southern doji) often do not work well as bottom turning signals. In this chart doji, or doji-like lines A to F, appeared during a market decline. But these Southern doji did not signal any reversals. The first sign of a turn came with the tall white candle on March 3 wrapping around the two previous doji lines at F. This formed a bullish engulfing pattern

Exhibit 6.14. Longview Fiber-Daily (Doji in a Box)

Exhibit 6.15. Owens Illinois-Daily (Southern and Northern Doji)

that held as support. (There was also a small bullish engulfing pattern at time frame C, but once the black candle closed under the low of that pattern, the hopes for a bottom were negated.) An indication that the rally from the bullish engulfing pattern at F was stalling came with the doji like line at 1.

In Exhibit 6.16,1 have a series of doji illustrating how different market surroundings influence a doji's importance. Let's look at each individually.

·       Doji 1. "An earthquake of magnitude 8" is how the Japanese would describe the action preceding doji 1. Two forceful black real bodies are followed by equally intense long white candles that made up all the lost ground of those black candles. The doji at 1 showed the stock was separating from its trend (in this case the trend was up because of the two long white candles). As is the case for a doji and tall white candle, we use the highest high of these two sessions (in this example, it is the high of the white candle at 3745) as resistance. This resistance held on the next session. 

Exhibit 6.16. NASDAQ Composite-5 Minutes (Doji)

·       Doji 2. This doji came after a short-term decline. (Even if there are a few down sessions before the doji, I still view the immediately preceding trend as down.) Therefore, as a doji in a downtrend, it warrants less attention as a turning signal.

·       Doji 3. This doji came after a tall white candle. As such, it does hint the rally that started from 3705 could be losing its steam. However, is the stock overbought when this doji arose? In my opinion, it wasn't (compare this to doji 1 that came after an intense vertical rally). Consequently, this doji has less meaning than it would in an overbought environment. Once the market closed over the doji, it negated any of its potentially bearish implications.

·       Doji 4. Doji 4 arose within a lateral price environment. Since there was no preceding trend to reverse (the doji was in a box range), it has less significance as a turning signal. One useful aspect of this doji was that it helped reinforce the resistance area defined by the bearish engulfing pattern built a few sessions earlier.

·       Doji 5. A Southern doji. Since it did not confirm any other bottoming signals, it is not important.

·       Doji 6. Same as doji 5.

·       Doji 7. This doji shows how candle lines and patterns must be viewed in the context of the price activity before that pattern. This doji came after a downtrend. Normally, as detailed with doji 2, 5, and 6,1 would not view a southern doji as a visual warning of a bottom. However, in the context of the overall market picture, this doji takes on more significance since it was confirming support. There was a bullish engulfing pattern at B at 3680 and the hammer on the session before doji 7 told us the market was building a base near 3680-3682. That is why this Southern doji at 7 takes on extra consequence, even though it comes after a decline. It was confirming the dual support area at the bullish engulfing pattern and hammer.

Let's look at Exhibit 6.17 to understand why doji should not be viewed in isolation. This doji following a tall white candle

Exhibit 6.17. AT&T-Daily (Doji at New High Close)

was also a new high close for the move. Doji or not, a new high close is a plus. Many traders use "line charts" that are based solely on the close. The media will say that the stock closed at a new high. Many Western indicators (moving averages, oscillators, etc.) are keyed on the close. As such, a new high close keeps the trend up. With this in mind, I normally suggest that if there is a doji, that is also a new high close for the move to wait for bearish confirmation of the doji. This confirmation would be a close the next session under the doji's close. In this example, we got that bearish confirmation with the next session's close. That black candle day was also a failure at our resistance area defined by the doji's resistance area near $45.50.