|
|
|
Bulkowski’s Tall Candle Support and Resistance
|
Market
Industrials (^DJI):
Transports (^DJT):
Utilities (^DJU):
Nasdaq (^IXIC):
S&P 500 (^GSPC):
|
As of 07/29/2010
10,467.16 -30.72 -0.3%
4,415.02 -5.30 -0.1%
387.34 -5.78 -1.5%
2,251.69 -12.87 -0.6%
1,101.53 -4.60 -0.4%
|
YTD
0.4%
7.7%
-2.7%
-0.8%
-1.2%
|
Tom’s Targets
 375 by 08/15/2010 |
CPI:  on 07/07/2010 |
|
|
Written and copyright © 2007-2010 by Thomas N. Bulkowski. All rights reserved.
Unusually tall candles (2 times the average height) show support or resistance 39% of the
time. Candles 4 times the average height do even better. However, price does not show a
tendency to reverse near the middle of a tall body.
Tall Candle Methodology
Test 1
I used 453 stocks of daily price data covering the 3 years ending May 28, 2007 in the first test and found 41,301 candles. Not all stocks
covered the entire 3 years. Each of the candles had bodies (open to close, not high to low) more than twice as tall as the prior 22
trading day average (approximately 1 calendar month). I then searched for minor highs or minor lows that peaked or bottomed within the body of the candle.
For candles with upward breakouts, where price first closed above the top of the candle, I searched for minor lows
and stopped the search if price closed below the bottom of the body. For downward breakouts, where price first closed
below the bottom of the candle, I searched for minor highs and the search ended when price closed above the top of the
body.
For example, in the figure, a candle with a tall body appears in mid March. Later, price forms a minor low that
stops within the body of the tall candle. The tall black candle had an upward breakout (price first closed above the
top of the candle) so I searched for a minor low.
Test 2
The second test was similar to the first except I required the candle bodies to be over four times the average height and also lengthened the search time
to the 15 years ending on the same date. Few stocks covered the entire period. I
found 25,285 candles that qualified.
|
|
Results
Test 1
- A body 2 times as tall as the average shows support or resistance 39% of the time.
- Black or white candles also show the same support/resistance rate, 39%. That is, 39% of the time, price reversed within the body of the candle.
- Reversals are evenly distributed over the candle body. In other words, price may reverse at the middle of the candle just as often
as any other area of the candle body.
Test 2
- A body 4 times as tall as the average candle shows support or resistance 66% of the time.
- Black candles show support/resistance rates of 67%.
- White candles show support/resistance rates of 65%.
- Reversals are evenly distributed over the candle body. In other words, price may reverse at the middle of the candle just as often
as any other area of the candle body.
Conclusions
Based on this research, it is unclear whether tall candles actually show support or resistance. Think of it this way. Imagine that a candle was as tall as your price chart.
In that case, every minor low or minor high would appear within the body of the candle. The success rate would be 100%. Thus, the taller the candle body the more likely it
will be that minor highs or minor lows will find their way into the body of the candle and reverse there.
Also, the idea that a tall body commonly shows support in the middle of the candle is probably incorrect, at least based on these tests. Price is no more likely to stop
near the middle of the body than anywhere else in the candle body.
-- Thomas Bulkowski
Copyright © 2007-2010 by Thomas N. Bulkowski. All rights reserved. How do I set a laser printer to stun?
|
|
. If you buy ANYTHING, they pay for the referral.
