I remember many years ago I started trading the line chart (they also showed support). Then after watching a few videos on Youtube I found out I was hopelessly outdated. I needed to change to candlesticks. Now after having traded years with cabdlesticks it strikes me. Do I ever really need them? Do I look at them and analyze “what is going on”? Are the bulls or bears in charge? A moving average will tell me that better than candlesticks.
I used to wait for the hammer at support. Or the other “good” variant, the shooting star. My conclusion was that they do nothing. Engulfing just the same. Morning star, gravestone doji, tweezer and spinning top. ■■■■■■■■ all of it. The only thing I can use is the Marobozu. When I see that as my entry bar, my hopes smplify a little. Also 3 bars with wicks after an uptrend very often signals the end of the uptrend. Anyone else feel they need candlesticks anymore?
I find candlesticks my best chart style, and so useful that I will follow what the candlesticks say in preference to my EMA’s. In fact I don’t look at any off-chart indicators at all.
I should say that the candlestick data most useful to me is the daily highs and lows so perhaps it would be more accurate to say I use bars than candles.
I have run some statistical analysis on candlesticks for 70 pairs and 5000 candles per pair across multiple timeframes. The chances of the next candlestick being up or down is 50/50 regardless of the previous candlesticks direction or pattern. There is a lot more data outside of the prices that affect the trading.
I’d agree that a single candle can’t have predictive edge on its own, separated from context. For example, a hammer at support means nothing by itself, but if it shows up right at an EMA confluence zone, after a long downtrend, with a sudden spike in volume, it’s reflecting a real shift in participant behavior at that specific moment, not because the candle shape itself has predictive power. Same goes for the Marobozu you mentioned, its value comes from showing strong momentum within that timeframe, not because it’s some “special” pattern
This tallies with my own observations in forex, though not as detailed and objective as your own.
However, I did carry out some historic chart-based obs on the US indices. Forex pairs tend to be mean-reverting and confined to long-term ranges, whereas the US indices tend to trend, most often upwards, for lengthy periods of time.
I found that once a consistent uptrend is well established, the frequency of days with higher daily Closes goes to 55%. This might not sound like a bg number - maybe not big enough as a trading strategy on its own - but it’s a sufficient advantage in risk reduction to be worth tactical development.
