do candle sticks really work?

Gambler's fallacy is tarders worse enemy.

Ron

 

What about Steve Nison's books? Are they good ? i.e; Can they provide useful information about candlesticks and their use?

 

Nicely stated.

It's interesting that you say "experienced" traders like those on ET are still struggling to find a viable method to extract consistent profits from the markets. What does that say about experience? The longer you stay in the game looking for an edge, the more money you will lose.

 

Sure they can, but they won't make you successful.

 

Trendline and Moving average are good at identifying trend.
Candle sticks and indicators are good for signaling trend change.
When trend goes on for a long time, candle sticks is usually the first signal for trend change. Sometimes it requires only one stick.
Let's see if candle sticks really work or not!

 

I lost faith in candlesticks one day when I decided to do a simple study that almost anyone could do on their own.

I downloaded a large amount of historical daily data for various equities from Yahoo! Finance for the purpose of examining the Bullish Engulfing Pattern, specifically. I wrote a program to run through the data. It extracted all bullish engulfing patterns that were preceded by 3 days that closed at lower lows (a downtrend, albeit brief). The program wrote the data into a spreadsheet, including OHLC data for the day that immediately followed the engulfing pattern. That day would be more likely to demonstrate a turnaround, so I thought.

In the spreadsheet, I selected my data range and eagerly drew my chart. I expected that I would certainly see that the day immediately following the bullish engulfing pattern would show a tendency to close higher. I was breathing heavily as I dragged my cursor across the screen--the excitement was palpable. Finally my chart plotted itself.

Never in my life have I seen a more perfectly drawn bell curve, centered precisely at 0. The closing price of the day following the bullish engulfing pattern appeared to follow a perfectly random distribution, not showing any bias one way or the other.

I suppose one could argue with my methods, specifically the way I defined a downtrend. But the disappointment at that point was great enough that I never repeated the experiment.

Humbly yours,

 

I'd like to know how p-a-o (price action only) traders detect over-bought/sold levels

without indicators.

 

I'm sure you would

 

There are so many holes with your little 'experiment' I wouldn't even know where to begin.

This experiment did not prove or disprove anything about candlestick trading.

 

There just can't be that many holes in it--it was just too straightforward. You can keep your religion or magical dojis and dark clouds or you can repeat the experiment yourself, if you have the skills.