September 22, 2011
To succeed as an investor you have to truly understand risk and divorce yourself from the destructive human emotions of greed and fear.
Technical Analysis
I had a really good question come in over the weekend from a reader. He asked about technical analysis and whether investors actually use this stuff.
Technical analysis is a way of valuing an asset. It focuses entirely on price action. The core curriculum of this school of investment thought is that price is the most efficient way of measuring the relationship between the supply and demand for a particular asset. The idea is that every external factor and every piece of fundamental data gets discounted into one price. Technical analysts look for patterns in the movement of this price, and from it, they draw conclusions about what future price action may look like.
There are a million guides out there on technical analysis. I’m not going to recreate them. I’m not going to assemble anything that Edwards & Magee haven’t already in their classic, Technical Analysis of Stock Trends. But this week I can give you a brief overview of what it’s all about, and more importantly, why people use it.
My dad was actually one of the early pioneers of technical analysis. I remember when I was a kid he used to lug around this big black chart book. I have vivid memories of vacations up at Lake Tahoe or Disneyland. He’d go out in the morning, call the office, and get all the price quotes or buy a Wall Street Journal. Then we’d sit down at the motel coffee table together and I’d read off the Open/High/Low/Close prices while he’d plot them in his book. It must have been an actionable practice, because sometimes we’d play Chinese Checkers after he finished charting and sometimes he’d have to go back out and make more phone calls. And it must have worked because that’s what paid for our stay at the Motel 6.
Technical analysis is basically religion. You believe in it or you don’t. The atheists won’t practice this stuff and no amount of convincing will change their mind. Technical analysis is nothing more than witchcraft to hardcore fundamental analysts. That’s OK. Everybody’s different.
But also like religion, there are varying degrees of faith within the segment of the population that does believe. There are extremists, and like any extremist, those that are slavishly devoted to technical analysis live lives that rational individuals have a hard time understanding. These folks also aren’t much fun at cocktail parties. They’re a little scary. And judgmental.
Most believers are somewhere in the middle. Some go to church every single Sunday and would never, ever buy a stock under its 200-day moving average. That’s perfectly sensible, and for these individuals, very comfortable. I’m a huge advocate for investors to find what feels right and roll with it. Others pick and choose the bits of the faith that are most interesting to them. Maybe they’ll only attend Easter or Midnight Mass and only occasionally short into a head-and-shoulders top. They’ll take the Lord’s name in vain and sometimes they’ll ignore the trendline. These folks are OK, too.
The thing that’s most important to understand about technical analysis is that:
- It is not The Answer.
- It is a spectrum of belief.
There’s a cliche in the investment world, “fundamental analysis tells you what to buy and technical analysis tells you when to buy it.”
I’ve always liked that. It’s not always or entirely correct, but I like that it weaves the two disciplines together in a manner that’s productive and actionable.
I do pretty much the same thing in my own investing. At the back of my mind I always have a list of assets I’d like to own. That list has been assembled almost entirely by fundamental factors. While I may like the assets, I may think like the prices are wrong. Technical analysis helps me figure out the prices at which I should be more active about owning. I feel like it helps me stack the deck in helping me find entry points with more favorable risk/reward characteristics. I’m patient about buying stocks and would rather do it when it’s trading down around key support levels.
What follows is a jargon-free guide on the foundation principles.
Keep in mind, though, that all this stuff ultimately comes down to faith. We each have varying levels of faith, and if there’s one lesson I learned from being raised Catholic, it’s that one’s faith is constantly tested.
The Ten Commandments of Technical Analysis
1. Trends
This is probably the most important concept from Technical Analysis. It’s the First Amendment. It’s Econ 101. It’s the basic tenet from which much of everything else flows.
Prices move in trends. They just do. There are a billion reasons why and they all work together to generate some type of trend. People have written huge books explaining them, but to my knowledge, there’s no single reason that explains why prices tend to move the way that they do. Trends are really easy to spot in hindsight, but they’re a little more difficult to identify in the moment.
Sometimes different flavors of trend may run counter to each other. For example, the stock market is currently in a secular bear trend. But the cyclical trend is bullish. And the very recent trend is bearish.
What it all means is subject to interpretation. Not all of the answers here are black and white. Some are more skilled at deciphering this than others.
Louise Yamada, one of my investment heroes and the High Priestess of Technical Analysis, probably knows what that chart means. She’s not the only one, but the reason why I listen to her specifically is that she allows for a range of outcomes. She might feel very confident and speak very authoritatively about the market moving in a particular direction, but she is neither surprised nor bent out of shape when it doesn’t. She understands that these concepts are guides, not rules.
Even if you don’t trade with the trend, it’s important to at least understand what the trend is. Markets behave in different ways when they’re trending in different directions, and understanding that can help you regardless of what strategy you employ or how you execute it.
2. Moving Averages
These are a way of quantifying the trend. They’re nothing more complicated than an average price for the last X number of days. Around here, we’re big fans of the 200 day moving average. 200 days is actually a rather arbitrary number. The point is that it’s a longer-term moving average. If you want to use a 193 day moving average, you have my blessing.
What we do is look at how current prices relate to the long term average price. If current prices are higher, that’s a sign that the trend is up. If they’re lower, it means the trend is down.
Here’s a 5-year chart of the S&P with a 200-day moving average. You can see that staying out of the market when it’s under the green line has saved you a lot of heartburn.
Watching something like the 200-day moving average certainly isn’t a risk management panacea. But it can help you identify environments where you may want to think twice about being long or short. It’s tough to fight the trend.
More advanced technical analysts will look at how different moving averages play off of each other. They might watch how the 50 day moving average moves about the 200 day. When one crosses the other, it’s an event that may have meaning. Some investors use those events as a buy or sell signal.
3. Support & Resistance
This is one of the most useful concepts that investors can monitor. It’s another thing that investors can easily spot on a chart and it’s easy to implement into actual trading. Support & resistance are the levels that act as roadblocks to the trend. Markets either punch through these roadblocks or bounce off of them.
That’s a 2-year chart of the S&P 500. You can see that the market tried really, really hard to get through 1350, but it just couldn’t do it. There’s probably a reason why, but within the framework of technical analysis, we don’t really care. We just care that it couldn’t. That was a meaningful event. Same thing with all that support down around 1050. Who knows why the market kept rallying off that level, but it did.
In this basic technical analysis study, one might conclude that it’s a good idea to buy around 1050 and sell around 1350.
This is where I use technical analysis in conjunction with my other research. If I have a fundamentally constructive view about the market, but lack conviction, I won’t do anything until the market gets down around that support level. I’ll re-check to see if my fundamental thesis is still sound, and then I’ll make a move. I might still be wrong on the investment, but the probabilities are more in my favor when I’m buying closer to support than resistance.
Markets are naturally drawn to levels of support & resistance. Right now, the market seems to have short-term support between 1125 and 1150. If the market breaks through there, that green line will start acting as a big magnet. The market will probably want to go down and retest that support.
4. Ranges
Basically what we’ve done with support & resistance is define a range. Markets trade in ranges.
Sometimes they trade in ranges between trendlines. Sometimes they trade between moving averages. And sometimes they trade between fixed price levels.
The thing about stocks (or any asset) is that they are valued based on a very long term stream of future cash flows. From a fundamental perspective, the present value of a company like Apple is calculated by discounting the expected streams of income it will generate in the future. As you can imagine, this is a very difficult, very imprecise way to come up with an exact price. ”Probability spheres” get involved and so do varying “discount rates”. Valuation can get a little woolly.
What we’re really left with in the present is a range of sensible values. And over the short run, a specific stock or the broader market will trade inside this range. Technical analysis just tells us how to measure it.
Again, technical analysts look at ranges to identify locations at which to buy or sell. They’re not a guarantee of a winning decision. But they help investors spot the points where the odds are tilted a little more in your favor.
5. Breakouts
When a market moves below or above support or resistance, you have what’s called “breakout.” Usually, it means that the market will keep right on moving in that direction.
Here’s a chart of a few breakouts in the VIX (volatility index):
I don’t ever trade off breakouts. But some people do. I just use them as an indicator to tell me when the game has changed and when I should re-evaluate my investment thesis. I use them in conjunction with other indicators and fundamental data points.
In the case of that VIX chart, the breakout above those old highs should have been a warning signal. Technical analysis was telling you that a major environmental shift was happening underfoot. On TV and from the fundamental camp, pretty much all you saw was shock and confusion.
In times of crisis and volatility, I frequently see investors turn to technical analysis for answers.
I think it’s the same mechanism that drives us back into church when we feel particularly adrift or afraid in our lives.
6. Oscillation
This is where it gets a little bit more complicated. We talked last week about how assets tend to trade inside a range. As it happens, there are ways that you can quantify that range. You can measure whether the current price is at a relative extreme.
I don’t want to take you guys too far out in the weeds with these indicators. The calculations can get highly technical. But they’re really cool and the basics are simple to understand.
Bollinger Bands are one of the easiest to use. They’re not technically an oscillator, but they are a way of mapping the back-and-forth movement of the market. The concept is simple. To draw some bands, all you have to do is plot a moving average — 20 days is very popular – and then draw a line two standard deviations above the average and another line two standard deviations below that moving average.
Here’s an example:
Bollinger Bands don’t generate very good buy/sell signals on their own. Other oscillators are much more reliable. But if you use Bollinger Bands in conjunction with other indicators, they do an incredibly good job of telling an investor when prices are relatively too high or low. In my own practice, I’ll use Bollinger bands to confirm the signals I get from other indicators.
Relative Strength is a personal favorite of mine. I’ve talked about it in this newsletter before. It’s one of the indicators I use to tell me when something is overbought or oversold.
I won’t trade entirely around it, but RSI is often the final trigger to get me to make a trade I’ve been watching and waiting on. I’m sort of a freak. I like to buy things that everybody hates and is desperate to sell. And I like to sell things that everybody loves and can’t get enough of. For contrarians and counter-cyclical consumers like me, oscillators such as the Relative Strength Index are a trusty tool to keep in your belt.
Here’s an example:
You can see that the RSI told you things were getting a little frothy back in late July. Maybe that was a good time to take some profits off the table. In mid-August, you can see that the market was pretty oversold. Maybe that was a time to scale back in for a little bounce.
Now, keep in mind that just because an asset is oversold doesn’t mean you want to buy it. And just because it’soverbought doesn’t mean you should sell. Markets do crazy things for crazy reasons, but sometimes they do very sensible things. Sometimes things are cheap and oversold for a reason. So be careful here.
Stochastics are kind of similar to RSI. The theory with stochastics is that prices tend to close near the highs in bull markets and near the lows in bear markets. When bull markets are getting tired, prices tend to close further away from their highs. And when bear markets are getting ready to reverse, prices tend to close further away from their lows. In other words, momentum is the first thing to change in a trend.
Here’s an example, with the oscillator below the chart:
If you’re really good at reading stochastics, you can use them to forecast reversals. In a simpler sense, they can also be used to identify places where the market is overbought or oversold.
The above gold chart is a good one because it demonstrates a couple of important points. The first is that just because a market has moved a long way in one direction doesn’t mean it can’t keep moving further in that direction. The oscillator was flashing “overbought” as early as July 11th. Yet the rally kept exploding for another month. The oscillator stayed high.
It also illustrates the important technical principle of divergence. Even though the gold price was going way up, the oscillator was slowing moving the other direction. That’s how you know when to buckle up for a reversal.
As always, these oscillators are best used in conjunction with one another and other technical and fundamental indicators.
7. Volume
I probably should have listed this as number one or two. Volume is a tremendously important principle in technical analysis. It’s one of the major pillars. It’s involved in almost every aspect of technical study.
However, I don’t spend much time studying volume. That’s just me. Because of all the automated, algorithmic, and high-frequency trading that takes place in the market, volume levels and patterns are completely different than they used to be for most of history. The staggering majority of all the trades transacted in a given day are so short-term in nature that they’re not useful. Most of what happens within the course of the day is noise, at least as it pertains to anyone with an investment horizon beyond next Tuesday.
I think of volume as an exclamation point, a bit of punctuation that modifies the sentence that came before it. If there’s a sell-off in the markets and the daily volume is very high, I interpret that as: today was a SELL-OFF! Volume amplifies the move of the day. If it’s a day in the market where the Dow goes up 200 points, but the volume is very light, I’ll read that as a general lack of conviction.
Volume also goes with the trend. In bear markets, most of the heavy volume comes on down days. In bull markets, the volume comes when the market rises. One of the reasons why I’ve been a little skeptical that this whole bear market that started in 2007 is over is because the volume profile has been pretty concerning. If the market starts making new and consistent lows on light volume, technical analysis tells us that that may be a sign that the primary trend is coming to an end. It may mean that sellers have finally exhausted themselves.
I should also mention that our firm has always had a very difficult time designing trading systems based around volume. Again, this is just our experience. We’ve found it to be a rather weak indicator, especially when designing strictly quantitative systems. Volume tells us very little about the future on its own.
8. Breadth
Market breadth is exactly what it sounds like. It measures the breadth of the market’s move up or down.
In its most simplistic form, you can calculate it by relating the number of stocks that rose to those that fell. If more companies go up than go down, breadth is said to be positive. If more companies go down than up, breadth is negative. Bull markets tend to feature lots of days where the market goes up but also a large percentage of individual companies go up. The opposite is true in bear markets. Remember that day in early August when the market crashed? It was a historically bad day, but to make matters worse, every single one of the 500 in the S&P went down. That’s the kind of thing an investor will see only a handful of times in his entire career.
You can fine tune breadth through technical analysis to make it a more useful indicator. This is what the Advance/Decline Line does. It measures market breadth over time. To calculate it, all you do is subtract the number of declining stocks from those advanced. Then add that to yesterday’s value. You can plot those points over time to identify trends in breadth, though this line has less and less meaning the further out on the timeline you extend it. Personally, I prefer to just monitor day-to-day levels in market breadth.
The McClellan oscillator fixes a lot of what’s wrong the the A/D line. I won’t go into the details of the calculation because it’s kinda complicated. Basically, it turns market breadth into a ratio that’s more useful to track over time.
Here’s a really nice chart:
I look for two things in the McClellan Oscilator. The first are sharp spikes up or down. These are “breadth reversals,” surges in the opposite direction of the trend. Often, these spikes will foretell extended advances or declines in the market. These moves are extra interesting if they also go against the direction of the market. For example, if market is going down but breadth has been improving, I might buckle up for a sharp little rally as the market catches up with the divergence in breadth.
The other thing I look at is the amount of time the oscillator spends above or below zero. By simply studying a price chart of the S&P in the first half of this year, one might have felt reasonably good about the health of the market. Stocks obviously took a hit after the Japanese earthquake, but they rebounded and kept testing new highs all the way into June. If you peeked under the surface, what you saw was a rather bad environment for breadth. Look how much more time this oscillator spent in the red in 2011. That’s more supportive of a bearish trend.
9. Patterns
Patterns exist. They exist in nature and they exist in the markets. Some patterns are easy to spot, while others are difficult. Some have meaning, others are arbitrary.
This is where technical analysis gets a little out of hand. There are something like eleven hojillion different patterns. I only know the really popular ones.
I’ll share a few of those with you now. We’ll use pictures.
Head & Shoulders Top (bottom)
This is the granddaddy of all chart patterns. It’s a rock solid pattern and its track record is quite good.
That picture says it all, and the most important thing to watch is the neckline. If that level of support is violated it means that the trend is probably over. This is a longer term pattern, and you tend to see it near the end of long, large trends in one direction.
I wish that picture included volume. Most technical analysis rookies ignore volume in head & shoulders patterns but volume is the critical element that makes this such a reliable tool. In the classic pattern, volume will be very high on the left shoulder, still kinda high at the head, but very low on the right shoulder. If you see relatively low volume on that third peak and the market heading towards that neckline, be prepared to run for the hills. Tighten up your stops.
Pennants
Pennants are pretty cool. Look for markets that are making lower highs but also higher lows. It’s a pattern that really stands out and usually it will break out dramatically in one direction or the other.
The trend is set depending on which direction the breakout goes.
Double Tops (and bottoms)
These are pretty self explanatory. Look for a market that tries and fails (usually dramatically) to break that first top. It’s another reversal indicator. It’s a sign that buyers are giving up on the market. The opposite is true for double bottoms. Those look like a big “W.”
10. Use multiple indicators together.
All of these tools have flaws. You can’t rely on just one of them. Investing with technical analysis is like building a house. You can’t build a house with just a hammer. But if you also use a saw and a screwdriver, you’ll get somewhere. Not every tool is required for every job. Some jobs require using multiple tools at the same time. The point is that the more proficient you are at using all of these tools, the nicer your house is going to turn out.
Use these technical tools with fundamental analysis, too. This can give you very powerful results, but for some reason, most analysts tend entrench themselves in only one school of thought.
If you see what looks like a head-and-shoulders top forming, check the relative strength. Is it overbought? Are the fundamentals of the asset kinda negative? Is the asset trading near the top of a short-term range? Has market breadth been getting worse? If you know how to do all of this stuff, you can really tilt the probabilities in your favor.
Obviously, you want your investment portfolio to be diversified. But you want your decision-making process to be diversified as well.
Here’s a pretty cool chart that a reader sent me. Last week it might have looked like total gibberish to you. But today I’ll bet you can make some sense of it.
This is a really good chart. You’ll have to click on it to make it bigger and see everything that’s going on. He’s looking at a lot of different things in conjunction. He’s looking at relative strength. He’s plotted out trendines. He’s identified various range channels and areas where there is support or resistance. He pointed out a recent head & shoulders top – with decreasing volume! He’s even looking at Fibonaccis.
Conclusion: History repeats itself.
I think this is why people use technical analysis. I’ve yet to meet a serious investor that believes the movement from day to day is random. So many of the things that have happened before happen again. The details aren’t always exactly the same, but people that are skilled at interpreting history are able to look through that and see the bigger picture.
I think this is why technical analysis works. Ultimately, these markets are driven by the actions and emotions ofhumans. It’s this same dynamic that powers all the major trends and inflection points in history. I don’t know why history tends to repeat itself. I don’t know what it is about all these humans that make them do the same things over and over again. I’m sure there are hordes of scientists trying to figure it out.
All I know is that they do.
Technical analysis is really just a way of codifying that principle and applying it to the investment markets. It’s a way of identifying patterns of behavior that investors have exhibited many times throughout history.
If you accept the premise that history repeats itself, then you’re implicitly accepting the premise that by studying history, you can divine a little bit about the future. If it happened this way once, then maybe it’ll happen this way again.
Now, I can hear you saying, “Hold on a sec there Mr. Nostradamus. It’s impossible to predict the future!”
This is both true and false. Consider some of these predictions:
- The 2012 presidential election will be an ugly fight.
- The Kansas City Chiefs won’t win the Super Bowl.
- The U.S. population will grow.
Keep in mind that those are technically predictions. Those events have not happened yet. But everybody knows that they will. So you can see that it is possible to predict the future.
The catch is that the more precise you try to get, the smaller the odds that your prediction will materialize. It’s easy to make those above three predictions about next year. They are virtual certainties. But it’s basically impossible to predict who wins the presidential election and by what margin, or what teams will make the Super Bowl, or exactly how many new American births there will be next year. I could make those forecasts, but the probability that they occur is very, very small. Predicting the future is all about probabilities.
Keep this in mind when performing technical analysis. These historical patterns can help you determine some broad generalities. But if you’re looking for exact moves on exact days, you’re bound to be disappointed. It doesn’t work that way. That’s the biggest thing that everybody gets wrong about technical analysis. It is not an exact science. Not even close.
The best that technical analysis can do is help you identify the places where the odds are tilted in your favor. Most of the time, the advantage will be slight, but if you do it enough and over a long enough timeline, all you really need is a slight advantage to make some significant money.
