Priority of Indicators

BobS

There are many things that a person should consider on a technical chart when considering whether the equity is about to change direction. Since often these indicators send different messages, it is helpful to have a general priority for them. I am interested in your perspectives on the relative priority of the following indicators, and any others which you consider more important than these. I am placing the following in the priority I am thinking at this time. Also, if you think that only a few of these are really relevant, and the others are comparatively insignificant, let me know that too.

Horizontal Support or Trend/Channel line (touch or cross)
Higher Volume on up days vs down days
Rsi Divergence
Cross of OS/OB on Stochastic
Top or Bottom of Bollinger Band
10/20 dma
Reversal candle pattern
Chart pattern (H&S, double top ...)
Fibonacci levels

Thanks in advance for your responses.

markd

Hi BobS, I like your question, although I'm not big on indicators myself. To me, indicators treat the market as if every day is the same, like the data is just measuring a stream of electrons - each data point gets equal weight. But the market is emotional - some bars matter more than others. Nevertheless, here's a list:

Trend direction (i.e. direction of long and intermediate term MAs)

Location of MAs (e.g. shorter above longer is bullish)

Pattern of highs and lows (e.g. failure to make new high followed by lower low)

Location of price vs:

MAs (above or below)
prior high volume bars (esp. price inside a HV or breakout bar range)
price channels (near - meaning abv/blw 80/20 on the K(x,1)s - or at up/mid/low chan)
fibs (usually for current and/or next larger leg)

Signs of trend strength vs. counter trend strength in price bars (candle form, range and volume)

Volume and candle configuration at locations of interest (signs of returning trend strength and counter trend climax or exhaustion).

Note: the list includes MAs, channels and stochastics, which might be considered indicators, but as used here, they are just a way of keeping track of where prices have been over time or where support and resistance are more likely to be found. They are not "signals" the way a MACD crossover is a signal.

BobS

Thanks MarkD. As usual your response was faster and more substantial than I expected. I appreciate all of the entries in your list and I understand most of them. Two of them I understand parts of but not all of. They are:

"prior high volume bars (esp. price inside a HV or breakout bar range)
Volume and candle configuration at locations of interest (signs of returning trend strength and counter trend climax or exhaustion). "

Taken together are these 2 points saying (in different words) that particularly HV bars are often at "locations of interest", especially at transition points like highs, lows, and breakouts?

If so, I understand much of these 2 points. I am still not real clear on what "price inside a HV or breakout bar range". Specifically I am unclear on the word "inside".

Thanks.
BobS

Kevo

Nicely crafted post. I believe the purpose of your list is to determine trend change. As I understand it, this is your priority list:

1. Horizontal Support or Trend/Channel line (touch or cross)
2. Higher Volume on up days vs down days
3. Rsi Divergence
4. Cross of OS/OB on Stochastic
5. Top or Bottom of Bollinger Band
6. 10/20 dma
7. Reversal candle pattern
8. Chart pattern (H&S, double top ...)
9. Fibonacci levels

Of your list, this is how I would arrange them:

1. Cross of OS/OB on Stochastic
2. Reversal candlestick pattern
3. 10/20 dma
4. Top or Bottom of Bollinger Band
5. Chart pattern (H&S, double top ...)
6. Horizontal Support or Trend/Channel line (touch or cross)
7. Higher Volume on up days vs down days
8. Fibonacci levels

Warning: This priority list depends on your goal/style/strategy/temperament.

I would scratch RSI divergence off the list. RSI(14) on a daily scale only becomes important if it has reached extremes - at or above 70, or at or below 30. On larger scales, I use RSI(14) to measure trend direction - meaning is it in a bull zone or bear zone. If neither, then expect volatility.

Reversal chart patterns are long-term and are best utilized on indexes.

If price crosses a short-term MA, then this is important. As markd points out, long-term and intermediate-term MA's are missing from the list for purposes of trend direction.

I think what markd is getting at is that the price on a very large volume day is significant. In other words, make a note of that price.

If the purpose of your list is to determine trend change, then I would make many more changes. This is not a complete list: Fist, I do not see momentum on this list. Second, external factors influence the security that indicators on the security alone will not catch. These external factors include the Industry Group, the sector, market capitalization, and overall market conditions like volatility, and market breadth. Third, instead of chart patterns, do peak/trough analysis. Fourth, scratch Fibonacci because that only helps to determine a retracement level. The important metrics to watch are: trend, momentum, overbought/oversold, support/resistance, accumulation/distribution, price relative, relative strength, and gaps. There are different ways to monitor these metrics. I usually include at least two of each category.

markd

Hi BobS, here's an illustraton of "inside a HV or breakout bar":



Oct 20 is an HV bar.

The 38 fib occurs inside (within) it's range, as does the 251MA. Price turns in it's range on 10/29. Price turns again on 12/16 after closing below the 10/20 low. At that time, the 251MC is within 10/20's range.

This is not meant as an example of a good trade, just the first chart I could find that clearly illustrates what you were asking.

As for "volume and candle configuration", I would read the bars leading up to the 10/29 low this way:
10/27 HV down bar with good range making good progress - strong assertion of down trend; close is within 10/20 HV bar range, but no sign of buyer strength
10/28 no follow through from sellers; buyers run up the price but on very low volume, so not much strength yet.
10/29 sellers return on high volume, but make small progress considering size of volume, esp vs. what buyers did on small volume day before, and buyers close prices off the lows; so despite down close, there is "hidden" buying absorbing the selling.

Also worth noting on this chart: the 38 fib occurs at the 10/8 low. That's a point in the prior decline where buyers were able to temporarily reverse the decline. So it's a price that is meaningful to buyers - it becomes support again on 10/29 and 10/30.

BobS

Kevo, thanks for noting that MarkD had pointed out that long-term and medium-term MAs were missing from my list. I will address another question to him about that. Also, thanks for listing the essential metrics to watch, namely "trend, momentum, overbought/oversold, support/resistance, accumulation/distribution, price relative, relative strength, and gaps".
I should note (something that I did not include earlier) that I am at this point especially focusing on index ETFs (like IWM, SPY, QQQ) and their leveraged versions. So, patterns that relate to indexes are relevant, as you have mentioned. Also, I do try to keep track of the price relatives between these. Also, I still have a lot to learn about gaps. I am in the early stages of learning about those. Thanks again.

BobS

Thanks MarkD. I have learned 2 big things for me to correct.

1. In your earlier post you talked about "returning trend strength and counter trend climax or exhaustion" and you also noted the simple point that when the intermediate MA was above the LT MA it was a bullish trend. Of course I knew this fact about the MAs. However, because I have always felt unable to make any money in a bear market (being wary of the risks of shorting) I have been trying to get a handle on using bear funds during counter-trend rallies in a bull trend. However, because of focusing on this, I have not treated these counter-trend rallies as that, but rather as the beginning of a new bear primary trend. So, my question on this is as long as I expect this counter-trend rally (in the bear fund of QQQ for instance) to last only a few weeks, is there danger in playing it for that period to get practice for the bear trend that is likely coming?
2. In your post this morning you shed MUCH light on the price and volume action between 10/24-30. Before your interpretation (and if I had not seen the action after Nov. 1) I would have looked at the huge volume red bars of that period and considered the far more impressive than the HV up bars of 10/14-20. And I would have been shocked to see it go back up starting Oct 30 on such meager volume. So, am I finally getting the point when I say that just huge volume on down days does not only mean that the bears have the upper hand over the bulls, but that there could also be "hidden" buying by very committed bulls, if the price action (that is low movement of price on big volume or tails indicating that bulls closed significantly off the lows) indicates as such?

markd

Hi Bob, I can't really answer question # 1 since I have never even looked at bear funds. But my attitude to the market is, it's always dangerous

On question 2, I think you're getting it, as you say. The point is, for the trend to advance (bull or bear) it needs to continue to show increasing strength in the trend's direction. It needs to attract followers (volume), and it need to make price progress commensurate with that volume (good range extending into new price territory, not just retracing a reaction). If it does something else, it's a warning.

So, for instance, why didn't bears turn up in droves after 10/27, a big down day? They had good reasons to press their advantage - the last big market move was a dramatic new low 9/22-10/15; the return rally ended on a very bad candle for the bulls 10/22 (gap up open, close at low) to make a potential lower high after a lower low; the bull response on 10/24 was weak; and then 10/27 the bulls were crushed. Yet on 10/28 there was no follow through. As it turned out, the bulls jumped on the hesitation.

On the other hand, it often happens that a trend will show weakness and the counter trend doesn't respond. You can have several warnings but the trend just keeps going anyway. Or, the counter trend might assert itself strongly, and the trend takes it as a wake up call and takes off on a new leg.

BobS

Question 1: So, in a period like 2008 would you just stay out or would you get into options or short?
Question 2: Since a trend showing weakness does not mean that the counter-trend will take over, is the reason why the bears did not push their advantage on Oct 28 & 30 remain a mystery? Or was it because the bulls had so many people in the wings who thought that area was a very reasonable price for XLB (judging from the price and volume action from 10/20, or was it the 38 fib level.

markd

1. You could buy puts or sell calls if you don't want to short; a bear fund might be the best way, I just haven't studied them.
2. Well, you can infer that there were not enough bears left to push the price down more. In other words, it looked like everyone who wanted to sell had sold. And you can infer that the bulls, seeing they would get less opposition than implied by the 10/27 deluge were willing to absorb the selling that did show up. But you never really know the true motives.

You could speculate that there weren't enough bears because we've been in a bull market for so long, there's no bearish crowd yet, so the conviction bears weren't yet been able to attract a me-too following. And you could attribute the strength of buyers to the die hard bullish crowd (including professionals) that sees a bargain in every retracement and just ignores even huge signs of seller strength (so far).

BobS

Yes, sentiment being extraordinarily high probably has a lot to do with it. Thanks.