I have developed my own custom scripts to help with my chart analysis so its not easy to answer some questions and everyone tends to develop their own favourite method and preferred indicators.
I think you find it useful to have detailed comments and to then go away and do some reading and study. Thus I will try to give a reasonably detailed answer rather than just a superficial one.
If I do a chart analysis without the aid of my own analysis scripts I probably first look at the overall shape of the long term chart and start from there. That gives an overall feel for the stock. I would then put on some of the main horizontal/flat support and resistance levels from the obvious major levels. Where there is a sustained movement in a particular direction (up, down or sideways) I would then probably look if it created a channel - which is what would normally happen for a sustained trend rather than a short term movement. Thus a trend that is sustained over a few months or more will generally have a channel. The price may have broken out of that channel at times but I can normally see a distant channel fairly clearly. Generally the main channel limits will connect multiple candles often including either highs (for upper channel limit) or lows. If you get it right you will normally see the price either hugging the line at times or bouncing off it If you look at the white dotted lines for the recent rise you will see the price hugging both lines particularly the upper line. The lines are just guides and act to mark the channel. They aren’t necessarily as fixed as horizontal/flat sr lines. They can help to indicate when the price breaks out of the channel but prices do often go outside upward or downward channels at times. In general price will stay in a horizontal/sideways channel much more and as a very general rule its probably more important to pay attention to breakouts in horizontal channels because they often either indicate a significant breakout (ie move to a completely different channel or direction) or are a fakeout (ie price moves in one direction as a fake move and then after a few days makes a strong move in the opposite direction). A break of a up or down channel can often simply move price into an identical parallel channel (ie direction is maintained just a slightly higher or lower channel - ie the long sustained rise I would say had two parallel halves to the overall channel and moved from one half to the other a couple of times during the overall rise. There are several classic continuation chart patterns where a channel breaks, price moves in the opposite direction for a while and then resumes the original direction. An example is the long sustained rise up to 330 and then a downward consolidation channel before resuming the upward trend. This was a classic continuation pattern.
So in summary - I tend to view flat sr levels as harder forms of support or resistance whereas upward or downward channels are a bit more flexible (ie even if the price breaks the channel limits it often either forms a parallel channel in the same direction or starts a consolidation pattern before resuming the original trend direction). Up and down channels do break and reverse but I think its more complicated than horizontal channels where the price tends to stay clearly within the channel until the channel breaks and those breaks can often result in strong moves up or down.
I do like moving averages. My own scripts use probably 12 or more moving averages of different types (both simple and more complex forms of ma). In the script I also use least square regression with multiple timeframes to find price direction and predict future price. I do think moving averages (and crossovers etc) are very good but my script does all of that for me now so I don’t tend to look at the individual lines much now but at the moment I am modifying this part of my scripts to add in calculation of multiple anchored vwap to 1) create further support/resistance lines and 2) for the script to use with the other moving averages to create a composite indicator which I do heavily rely on. The way I calculate this composite indicator creates a highly sensitive indicator (ie responds within 1 chart bar to indicate potential change in trend but has some mathematical filters applied to reduce false signals and create clear buy/sell signals. As a general guide I think moving averages are very good especially if you look at the crosses (ie one ma moving over another) and look at whether price is above/below or crossing. The trick is finding the right time periods because if you look at very long time period ma it is a very lagging indicator which isn’t good for some stocks that don’t have sustained trends whereas normally fast moving ma can be too sensitive and generate too many false signals. Also ema or other forms of moving average will be more responsive than sma but sma are good for overall trend.
I don’t use bollinger bands or macd. I know some people are fans of both but I have never got on with them. I do like RSI and Stochastic but don’t use either now because my scripts contain highly modified forms of them to overcome what i felt were weakness in both rsi and stochastic. However, even in their standard forms I think rsi and stochastic are both good indicators.
I also use Williams Fractals, Heiken Ashi and supertrends and my script calculates forms of all of these to combine with the other indicators. Individually they all have strengths and weakness so its a question of how you combine them or analyse them - ie what you use to confirm indicator signals.
I do think there is a difference between what you can reasonably do as chart analysis by just looking at a chart and what is possible by writing code or a script to analyse it all mathematically. Thus something like Heiken Ashi will have several theories on how to get confirmation or determine buy/sell signals but I think that a script can be far more complex and subtle. Nothing is perfect and scripts takes a massive amount of refinement but can potentially be far more complex and look at a wide range of factors.
The actual candlesticks I look at somewhere in the middle of the analysis and at the end. Candlesticks by themselves are not very reliable indicators but they can be good at confirming other aspects of the analysis. With moving averages and support/resistance the candlesticks are fairly powerful at showing how the price is interacting with the ma or SR levels. With the other indicators (ie rsi, stochastic, etc.) I think the candlestick give more of a confirmation or double check
So my basic list is:
- flat support resistance
- inclined channels (support resistance lines)
- moving averages
- candlesticks (ie look at how price interacts with ma or sr lines)
- rsi and stochastic + some others
- candlesticks to confirm the overall analysis
As I say everyone has their own method, own preferred way of doing chart analysis and own preferred indicators. There is no one perfect way or solution. Its all personal choice
