I've been hanging around with people who disparage technical analysis. I've been using technical analysis less than I used to in the past, but, on reflection, that's not really the fault of the new crowd.
I'll give you a little background. I had just turned 50 when I began actively trading, but I prided myself on my forward-looking view of the markets. I investigated every new technical analysis tool my charting platform provided. I loved technical analysis because it brought back the kind of studies I had found fascinating when I was an eighteen-year-old physics major at a small university. I trusted math, and much of technical analysis is about the mathematical manipulation of various inputs and then the charting of those results. Some technical indicators are related to ballistics, which comprises a subset of physics. I was in my element.
Not all of technical analysis requires a mathematical manipulation of data. Sometimes technical analysis requires pattern recognition. That kind of pattern-recognition skill set also appealed to another part of my brain that had once so loved math and physics. I was so enamored of technical analysis that I layered on four or five different lower indicators on my charts. I considered following the example of my colleague on the site, Jane Fox, and moving to a charting platform that allowed me to program in my own signals, particularly as I wanted to watch Donchian channel breakouts, a special interest at the time.
However, I never got to the point of learning the programming language I would need or switching platforms. I started experiencing something that many other amateur technical analysts experience. That "something" was signal overload. What if the indicator evidence contradicted what I was actually seeing on the charts? For example, what if indicators all pointed skyward but weren't yet at levels that typified overbought levels, but all the candles on my 30-minute charts had spiky upper shadows that pierced resistance while the candle body formed below it? What I was seeing on the chart itself was whispering warnings to lock in profits. But should I ignore the evidence of all those layers of indicators, precise and mathematically-based indicators that were telling me the trade still had some juice?
Many times, yes, I should have ignored the indicators and locked in at least partial profits when I saw evidence that there was some selling into price spikes, the action that created those spiky upper candle shadows. My observations resurrected an old concept from my high school and college days: collinearity. My reading taught me that many technical indicators and overlays were collinear. This means that they were not always independent studies, but were often in fact just derived from a different manipulation of the same data. A big group manipulated data related to price. All those were collinear. Another group might manipulate data related to volume. Those were collinear to each other, but not collinear to the ones that manipulated price. Another group might combine the two.
I could layer as many collinear indicators on the bottom of a graph as I wanted. They weren't going to give me independent data. One might signal a little earlier than the other and one might signal a little later but prove more reliable. There were slight differences, but those that manipulated the same data weren't giving me independent evidence of what was going on in the markets.
I found myself gradually dropping the number of upper and lower chart indicators and signals I used. Along with wanting any indicators I used to be truly non-collinear or independent of each other, I also was beginning to change the way I used technical indicators and overlays. I gravitated away from using them to provide trade signals and toward using them to help me prepare what I call "what-if" plans. As I navigated the two recessions and a number of massive, won't-stop rallies we've had since 2000, I stopped thinking I knew where the markets were going and began thinking in terms of the markets being "vulnerable to" a move to certain higher or lower prices. In addition, my method of trading had changed from pure directional trades to income-type trades in which I was selling premium and setting up positive-theta trades that would benefit from the passage of time. What I needed to know was whether the market looked vulnerable to a move out of the range that would bring me profit. I didn't need hair-trigger indicators any longer, with the possible exceptions of guiding me to the best entry on my trades and warning me when I might need to hedge my risks.
So, what chart setups do I use now? I'll show you. Remember, however, that my articles are roughed out a couple of weeks before publication, so the charts are not up to date.
Annotated Daily Chart of C:
Isolating each of these on the chart will help pinpoint how and why I use them.
Annotated Daily Chart of C with Candlesticks:
Annotated Chart of C with Keltner Bands:
[Note: The next trading day after this chart was snapped, C did climb to a high of $3.56 and closed at $3.54, but it couldn't maintain those closes above $3.40. It has not yet dropped to $2.81 but did reach a low of $3.14 on 1/27. C has, however, closed back above the black channel's support and the red 9-ema, so evidence is still mixed about where it might go next.]
Additional information that Keltner bands might provide is evidence that a move could be way overdone, as occurred when C breached the upper purple channel in August. It can tell me when an original supposition is being proven wrong, as when C broke through black-band support in December and appeared to set a potential downside target that was then near $2.63. When C scrambled back above the black band and began forming daily closes above that, that potential downside target was negated for the time being.
The settings I use for Keltner bands are 120-ema and 7.2 offset for the larger band, 45-ema and 3 offset for the black band and 9-ema and 1.4 offset for the smallest blue band. Some traders use Bollinger bands the same way I use Keltners. I'm just more familiar with Keltners and prefer them, but there's nothing wrong with another channeling system. The point is to become familiar with the way one's preferred underlying behaves with respect to the chosen channeling system.
Annotated Daily Chart of C with OBV:
OBV is on-balance volume, an indicator that Leigh Stevens recently covered in one of his Trader's Corner articles on technical analysis. As Leigh mentioned, this indicator gives quick visual clues about whether a stock is being distributed (sold) or accumulated. Sometimes this volume indicator can provide information that controverts what's seen in either the price bars or price-based indicators, although it didn't give much new information on C's chart. When OBV didn't roll over after the AUG 7 candle, the first red candle at the left of the chart that had a long upper shadow, I might have been warned that there was still some money going into C on balance. If I'd been long C, I would still have kept my what-if plans in place, but I would have been alerted that it might not yet be time to plow into a bunch of puts. Right now, it's not giving me much information.
Annotated Chart of C with RSI:
I speak of bears being prepared for an entry, but I don't know that I would have chosen such a low-price stock for my bearish trade, if I were into directional trades. I'm not offering suggestions for specific trades nor specific trading vehicles, just using these charts to illustrate my preferred technical analysis tools.
In addition to watching RSI's direction and absolute level, I also find it helpful to watch any chart formations that might be apparent on RSI and CCI. Both tend to signal earlier than other indicators, but their signals can also be false ones. So, as I do with all technical analysis tools, I use RSI to warn me to set up what-if plans to be instituted if the price action follows through. I don't use RSI as a signal, but as that warning to firm up my plans.
As this article was prepared, RSI had headed down and broken through a rising trendline formed since the middle of December, just as it had broken through similar rising trendlines in September and November. RSI had formed each of those rising trendlines as prices formed what looked like rising bear flags. RSI breaks of its rising trendlines either preceded or accompanied price breaks below rising support. Watching these formations would have alerted those in bullish position to cinch up their stops and would have alerted bears who wanted to get in on breaks below the bear flag support to be prepared.
Each of those previous breaks of bear flag support led to a lower price low on C. Would it this time? I thought we might know by the time this chart appeared, but all we know for certain yet is that C did follow RSI down, but only to an essentially equal low. It has not yet produced a lower low, but there was only one cent's difference. So far, RSI/price comparisons at last week's low have produced tentative bullish divergence, but such divergences are not necessarily trustworthy in a trending down market, as C has been in since last fall.
The study of Keltner bands gave me another potential price level to watch on daily closes to verify any breakdown, with that potential Keltner target not having been reached at the time this article was written and with the target being called into question by C's bounce back above black-channel support. Clues are mixed, so if I were thinking of trading C either direction, I likely would have stood back.
The information I garnered from the study at the time the article was roughed out would have stayed my hand then, too, if I'd been considering entering any particular trade as the information was not clear. Adding more indicators would not likely have clarified the picture, either. If I had added another indicator, it likely would have been MACD. I also like MACD because it does not have a limit of 100 or -100, as does RSI, so it's easier to see bearish or bullish divergences at extremes.
I like the RSI's earlier signals even if the trade off is that those signals are less reliable and RSI also offers less reliable divergence information. That's okay with me because of my style of trading. I just want to be warned to spiff up my trading plan or perhaps hedge risk. I'm not seeking information on when to enter a day trade. Someone else might make a different decision.
Learn what combination of indicators and chart setups work best for your style of trading and for the vehicle you trade. For example, if you're a trend trader, look for underlyings that have established a trend and are climbing or falling. Set a chart up with important moving averages for your underlying. Adding a drawn trendline and perhaps MACD might constitute your preferred setup. Traders in trending markets would use something like MACD only to confirm in the direction of the trade, not to give a fade-the-trade signal. Oscillators don't tend to be as reliable in fade-the-trade type signals during trending market behavior. A trader who elects to enter a new position or add to one with that convergence of signals at least has an advantage in having a close stop if the trade goes wrong right away.
A trader who trades channeling stocks might like stochastics. Someone like me who trades a lot of iron condors or other trades that depend on the underlying staying within a certain range might want to check ATR for that underlying or calculate a standard deviation move.
There's no right or wrong, a theme you'll hear from me a lot, but many of us gravitate toward fewer indicators or overlays, gaining familiarity with the ones used. So, while I like to play around with each new indicator my charting program provides, I seldom am swayed to add more indicators.
Despite the disparaging remarks I sometimes hear about technical analysis, I believe in its usefulness as long as we understand its limits. If we can't control the markets, neither can a Keltner band or a MACD signal. Both, however, can help us set up appropriate entries and stops in case our market analysis is wrong. They can alert us that something out of the ordinary is happening, so that we can stay out of the markets entirely.
Leigh Stevens covers many technical indicators in addition to the OBV, going into depth about how they're used. Be sure to check out his articles.
May all your indicators be true to you!