Are intraday Candlestick patterns useless?
- Thread starter newguyintown
- Start date
Hmm, I think 90% of candles are useless, but some 10% can be a goldmine,
Stacy
Stacy
There is no "main crux behind candlesticks". Candlesticks merely compress price data in a way that makes it easy for human beings to see range and time in a continuous time series. The only investor sentiment behind them is yours. It is you and your eyes that see the significance, the patterns, in the data. However, the fact that you think you see a pattern does not give it inherent meaning.
Your logical mind has concluded quite correctly that there is no significance to the samples of price data gathered at 15 minute intervals, nor those samples designated as "close", "open", "high", "low". These are terms applied to arbitrary numbers in a continuous time series. The terms imply meaning but have none, because they are not relevant to the data under study. I find it interesting that you have already made this observation yourself, yet you continue to maintain that there is significance to the same terms and numbers when sampled at a longer interval. I would submit to you then, that your obvious desire for the candlestick patterns to have significance is affecting your reasoning.
In the context of market microstructure, it is true that there is some significance to the daily open and close prices of the various markets. It is because different markets have special conditions that apply at the open or close of the market only. In the US equity markets for example, significant institutional volume often trades at the opening and closing cross. Sometimes these prices and volumes can represent a temporary imbalance of supply and demand, but that does not implicitly mean that they reflect sentiment, or that they are predictive of future prices.
No. It negates the significance of patterns, and candlesticks, in general. You are just as well off using chicken bones to decide where to drill for oil.
===========
New guy in town;
Like them better than bar charts, William O Neil prefers daily-weekly bar charts;
however thats personal.
Another point, its easier, buit not easy , to trade on the probability that SPY,ES, DIA,YM daily- weekly,daily trend is up;
but see what happened yesterday afternoon.-countertrend-intraday.
In other words, any market can be more difficult with smaller time frames & add MORE slippage, comissions;
a 1 minute candle charts & tick charts does me more harm than good, with my personality.
60 minute candles/bar charts can be quite helpful-profitable;
if worked 72 hours a week or so.
Medical DR requires 4 years college -school;
than 4 more years med, add more for intern, then money pours.
newguyintown,
In your first post, I was able to detect quite readily that you have a psychological preference for candlesticks. That is, you want candlestick patterns to be real , and you want them to work for you . NihabaAshi made a lengthy post with lots of detailed information about how he uses candlesticks, and it is clear that you have significant appreciation for this information. To me, it seems like you are being affected by confirmation bias, or a tendency to seek and use information that confirms your hypothesis or preconception, to the exclusion of disconfirming information.
http://en.wikipedia.org/wiki/Confirmation_bias
-segv
"Another point, its easier, buit not easy , to trade on the probability that SPY,ES, DIA,YM daily- weekly,daily trend is up;
but see what happened yesterday afternoon.-countertrend-intraday."
however, other signals CLEARLY indicated a reversal yesterday
1) the DOW had 7 up days in a row. the dow (unlike a coin) is not random. prices, and traders have a memory. it is rare (even in a roaring bull market) to get 7 up days in a row. so, i entered the market with a preconceived short bias. i took some long scalps, but waited for a short to present itself
2) the market reversed off of a daily pivot level.
3) market internals were WEAK (A/D and sectors)
4) Banks and brokers were not participating (although SOX was) in the up move)
5) the DOW had an open gap on the daily from several days ago. Open gaps on the daily that don't get filled intraday are strong magnets
6) the dow showed negative divergence when it made it's daily hi
etc.
i'm not trying to sound like nostradamus here. i am saying that candles are just a SMALL part of the analysis.
i sold close to the high and covered my last car at the gap. (11300). it was one of my best trading days ever.
candles just model price. they are not a trading system, imo.
in regards to the arbitrary time periods, this is what i look @
5 min chart
3/2 min chart (i switch back and forth on this window)
89 tick chart
i will also look sometimes @ 1 min chart at certain key price levels.
and of course these different candles will often show different, or even contradictory "signals".
as will ALL SORTS of indicators. the only indicators that consistently show the same signals are indicators that are - essentially - the same
thus, if two oscillators show "overbought", that is not confluence.
that is merely confirmation that they are both oscillators.
otoh, if you see confluence between a candle pattern, a key price level, a tick extreme, and an oscillator divergence
that would be a high probability trade
but see what happened yesterday afternoon.-countertrend-intraday."
however, other signals CLEARLY indicated a reversal yesterday
1) the DOW had 7 up days in a row. the dow (unlike a coin) is not random. prices, and traders have a memory. it is rare (even in a roaring bull market) to get 7 up days in a row. so, i entered the market with a preconceived short bias. i took some long scalps, but waited for a short to present itself
2) the market reversed off of a daily pivot level.
3) market internals were WEAK (A/D and sectors)
4) Banks and brokers were not participating (although SOX was) in the up move)
5) the DOW had an open gap on the daily from several days ago. Open gaps on the daily that don't get filled intraday are strong magnets
6) the dow showed negative divergence when it made it's daily hi
etc.
i'm not trying to sound like nostradamus here. i am saying that candles are just a SMALL part of the analysis.
i sold close to the high and covered my last car at the gap. (11300). it was one of my best trading days ever.
candles just model price. they are not a trading system, imo.
in regards to the arbitrary time periods, this is what i look @
5 min chart
3/2 min chart (i switch back and forth on this window)
89 tick chart
i will also look sometimes @ 1 min chart at certain key price levels.
and of course these different candles will often show different, or even contradictory "signals".
as will ALL SORTS of indicators. the only indicators that consistently show the same signals are indicators that are - essentially - the same
thus, if two oscillators show "overbought", that is not confluence.
that is merely confirmation that they are both oscillators.
otoh, if you see confluence between a candle pattern, a key price level, a tick extreme, and an oscillator divergence
that would be a high probability trade
I recommend either getting a good data provider or getting a backup data provider hooked up on a different internet connection for confirmation to the primary data provider if someone is serious contemplating getting involved in Japanese Candlesticks when just a few tick difference can determine if your have a confirmation signal or not.
Yes...having data problems once in awhile is normal (internet isn't 100% perfect every trading day) but if someone is having chronic problems they need to make changes or don't use Japanese Candlesticks.
Hmmm....interesting.
I was under the impression that he had doubts about specific situations.
Whitster has a sentence that sums it all up...
I guess I also have confirmation bias to what whitster said.
Then again...maybe we are getting too deep here.
Mark
(a.k.a. NihabaAshi) Japanese Candlestick term
Quote from maxpi:
So true. I experimented once with that and I was able to change the color of a whole series of candles [up bars to down and vice versa] in a sideways market with a small time offset.
If your data is giving you incorrect candlesticks, then it will also give you incorrect bars and incorrect data on any type of chart. Either way the chart is useless. Garbage in = Garbage out.
So true. I experimented once with that and I was able to change the color of a whole series of candles [up bars to down and vice versa] in a sideways market with a small time offset.
If your data is giving you incorrect candlesticks, then it will also give you incorrect bars and incorrect data on any type of chart. Either way the chart is useless. Garbage in = Garbage out.