technical analysis in the 21st century

A reader asked last week what I think about technical analysis.  This is my answer.

what it is

Technical analysis in the stock market is the attempt to predict future stock prices by studying current and past patterns in the buying and selling of stocks, stock indices and associated derivatives.  The primary focus is on price and trading volume data.

Technical analysis is typically contrasted with fundamental analysis, the attempt to predict future stock prices by studying macro- and microeconomic data relevant to publicly traded companies.  The primary sources of these data are SEC-mandated disclosure of publicly traded company operating results and government and industry economic statistics.

what the market is

The stock market as the intersection of the objective financial/economic characteristics of publicly traded companies with the hopes and fears of the investors who buy and sell shares.  Fundamental analysis addresses primarily the companies; technical analysis primarily addresses the hopes and fears.

ebbing and flowing

To be clear, I think there’s an awful lot of ridiculous stuff passing itself off as technical “wisdom.”  The technical analyst’s bible (which I actually read a long time ago), the 1948 Technical Analysis of Stock Trends by Edwards and Magee, is now somewhere in my basement.  I’ve never been able to make heads nor tails of most of it.

On the other hand, in the US a century ago–and in markets today where reliable company financials aren’t available–individual investors had little else to guide them.

the old days–technicals rule (by default)

What individual investors looked for back then was unusual, pattern-breaking behavior in stock prices–because they had little else to alert them to positive/negative company developments.

I think this can still be a very useful thing to do, provided you’ve watched the daily price movements of a lot of stocks over a long enough period of time that you can recognize when something strange is happening.

the rise of fundamental analysis

Starting in the 1930s, federal regulation began to force publicly traded companies to make fuller and more accurate disclosure of financial results.  The Employee Retirement Income Security Act (ERISA) of 1974 mandated minimum levels of competence in the management of pension plan assets, laying the foundation for the fundamentals-driven securities analysis and portfolio management professions we have today in the US.

past the peak

The rise of passive investing and the rationalization of investment banking after the financial crisis have together reduced the amount of high-quality fundamental research being done in the US.  Academic investment theory, mostly lost in its wacky dreamworld of efficient markets, has never been a good training ground for analytic talent.

The waning of the profession of fundamental analysis is opening the door, I think, to alternatives.

algorithmic trading

Let’s say it takes three years working under the supervision of a research director or a portfolio manager to become an analyst who can work independently.  That’s expensive.  Plus, good research directors are very hard to find.  And the marketing people who generally run investment organizations have, in my experience, little ability to evaluate younger investment talent.

In addition, traditional investment organizations are in trouble in part because they’ve been unable to keep pace with the markets despite their high-priced talent.

The solution to beefing up research without breaking the bank?  Algorithmic trading.  I imagine investment management companies think that this is like replacing craft workers with the assembly line–more product at lower cost.

Many of the software-engineered trading products will, I think, be based on technical analysis.  Why?  The data are readily available.  Often, also, the simplest relationships are the most powerful.   I don’t think that’s true in the stock market, but it will probably take time for algos to figure this out.

My bottom line:  technical analysis will increase in importance in the coming years for two reasons:  the fading of traditional fundamental analysis, and the likelihood that software engineers hired by investment management companies will emphasize technicals, at least initially.





Tesla (TSLA), me and momentum investing

Why should a company fundamentals-driven investor have a problem with momentum investing?

Two reasons:

–momentum investing is a reactive strategy, and

–one that focuses son the past price movement of the little pieces of paper (or electronic impulses) that trade in the secondary market.

In contrast, fundamental investing is a predictive strategy based on the idea that the price of the paper/bits will ultimately be determined by the value of the underlying company.  Among fundamental investors, value investors believe that the key is the worth of the company as presently constituted (but perhaps running more smoothly than it in fact is).  Growth investors think the key is in early recognition of novel and unexpected profit positives that will fully emerge only in the future.


What kind of a thing, reactive or predictive, is my formula for TSLA of:   buy at $180 and sell at $250?  In a sense, I’ve got some fundamental underpinning.  My back-of-the-envelope figuring suggests nothing is likely to happen inside the company Tesla over the next couple of years that could possibly justify more than a $250 price.  And I’m willing to sell at that price even though the stock is still exhibiting positive price momentum.

But how did I get the $180?

What I’ve really done is to take a chart of the stock and draw a line that runs through the lows of the past four years or so and to conclude that this line forms the bottom of a channel (with something like $250 as the top) that TSLA has been navigating itself through since late 2013.  Yes, at $180 I have better potential for upside than I do at $250.  But that’s more a fact about arithmetic than a deep insight into corporate operations at Tesla.

In sum, then, the fundamental underpinning of at least the buying are pretty lame.

So I guess I have to say that there’s a healthier dose of momentum in my fooling around with TSLA than I might like to admit.  On another non-fundamental note, though, this ensures that my California son and I stay in regular contact.

D0w 20,000

The question is whether having a badly constructed, information-poor stock market index achieve a round-number milestone has any significance.

It’s no secret that I’m not a fan of the Dow.  But I am of two minds:

–On the one hand, the Dow Jones Industrial Average, which is what we’re talking about, has been around for a very long time.  It has somehow deflected all attempts over my investing career to replace it in the media with an index that’s more useful, like the S&P 500.  So there’s at least a vague association in the public’s mind between Dow achievements and general economic prosperity.

–On the other, the minute you hear a media pundit use the Dow to interpret the ebb and flow of stocks in general, you know he’s clueless.

The best I can say is that if the Dow can remain higher than 20,000, doing so will have some small positive psychological value.  So I’ll take it.



Dow 20,000, S&P 2260

I’m not a fan of the Dow.  It’s a weird index whose main virtues are that, way back when, it was the financial media’s first try at measuring the US market and that, despite its peculiarities, it’s easy to calculate.  It’s no longer a useful gauge of US stocks, however.  So it’s never used by professionals, only by media people who have little industry background.  (One caveat:  the Dow indices are now controlled by the same people who own the S&P–who now have a vested interested in keeping the Dow alive, despite its drawbacks.)

Still, it’s striking that for the past six weeks 20,000 on the Dow has shown itself to be a strong point of resistance to the US stock market’s upward movement.  The equivalent figure for the S&P 500 is 2260, not a memorable number.

Whether the resistance level is 20,000 or 2260 makes little economic or financial difference.  Psychologically, however, 20,000 is much more daunting, I think, than 2260.  This is especially so now that the US stock market has risen far above former highs.

My bottom line is that, whatever number you choose, the post-election rally has run into its first substantial roadblock.  It’s also at least thinkable that the Dow is developing, at least for the moment, more relevance than I’m willing to give it credit for.  This would suggest that the balance of market power is shifting away from professionals to individual investors who have little stock market experience.    I find this hard to believe, but it’s something I should keep an eye anyway.


what I find most surprising about Tesla (TSLA)

a concept stock

My California son got me interested in TSLA a couple of years ago.

It’s a “concept” stock.  That is, the stock trades on the dream or vision of future revenue and profit.

…like Amazon

In many ways, it’s like Amazon (AMZN) was in the late 1990s.

That company seemed to me to be on the verge of financial disaster for most of the first decade of its existence.  It only began to be profitable after it expanded from its original virtual bookstore idea to becoming an online department store.  In my view, had AMZN not aggressively raised a lot of capital during the Internet Bubble, it would not have survived.  After all, it lost money eight (?) years in a row before breaking into the black.

the center of an empire

TSLA is the seat of the Elon Musk empire.  Some say it’s a car company (me included); some would characterize it ultimately as a battery company, with cars as the wrapper that contains the principal TSLA product.

the stock

The stock is now trading at $260 or so a share, giving TSLA a market capitalization of about $39 billion.  Suppose we think, to make up a number, that the stock should trade at 30x earnings.  If so, the current price expresses investor belief that at some point the company will be making $1.3 billion a year and still have, say, 20% growth in annual profit in prospect.

back of the envelope numbers

Let’s say TSLA is a car company and that it will be making on average $7,000 a car, after tax, on its output at some future date.  If so, the current market price already factors into it that TSLA will be selling about 200,000 cars a year–and expanding rapidly.

I think that’s possible.  More important, the market says that’s what investors are willing to believe, and pay for.


There are risks, yes, the most obvious of which is that the company keeps pushing back the date when it will turn cash flow positive.  What cash flow positive means is that the company will be able to generate enough cash from operations to cover costs, and will no longer be eating into its cash reserves to make ends meet.

what I find surprising

What’s stunning, though, is that less than two months ago the stock was trading at just over $141, or just over half today’s price.

New information has come out since then:

–TSLA began taking deposits for its $35,000 base price Model 3.  In less than a week, it has collected $1,000 each for about 300,000 units, with enough add-ons to bring the average selling price to $42,000. Most won’t receive their cars until 2018.  This support seems to me to show there’s potentially huge demand for electric cars, even at today’s lower oil price.

–the company announced that it missed its 1Q16 sales target because of parts shortages.  Presumably this means it did not turn cash flow positive as anticipated during the quarter.  That’s bad, especially since we’ve heard this song before.

the stock price

The stock is up $10-$20 a share on the two items, which were announced at roughly the same time.

What I find interesting is that a relatively large market cap company can move from $140 to $240 in a matter of weeks on a change in sentiment.  That’s about 70%!

So much for efficient markets and investor rationality   …not that anyone outside the ivory tower believes in this stuff.  But this is a huge move.

algorithmic trading?

I think it’s evidence of relatively naive algorithmic trading at work (based ultimately on two other wacky academic ideas–that the most important thing in investing is to control costs, and that there’s no craft skill/specialized knowledge involved in investing).

I also see it as support for my view that trading can be unusually profitable in this environment.   We should look for other instances where this may be happening.




the last bull standing

Being the last bull standing isn’t necessarily a good thing.  The Wall Street cliche is that the bear market doesn’t end until the last bull capitulates.  The complementary, equally hoary, standby is that the bull market isn’t over until the last bear capitulates.

To my mind, both are useful guidelines but not infallible ones.  On the one hand, markets are inherently cyclical.  So if an equity investor has a close to infinite capacity to endure pain–and if his clients don’t fire him in the meantime–persevering with a bad-performing portfolio can eventually pay dividends.

More qualifiers:

–that’s assuming the companies whose stocks the suffering-oriented manager holds are inherently sound, and not barreling down the road to bankruptcy.  They’re just poorly positioned for the current economic environment, which will sooner or later change for the better.

–a surprisingly large number of pension clients love to hire managers who have a recent hot hand and to jettison ones who are cold as ice, no matter what the long-term record.  So my “if his clients don’t fire him” qualification is a much greater risk than one might think.

On the other, there is something to the idea that until the most vociferous and publicity seeking defenders of a given position that’s going wrong give up, the situation rarely changes.  This may be as simple as that when, and only when, the buyers of what short-sellers want to offload disappear, so too does the short-selling–and hence the downward pressure on the stock in question.

…which brings me to Valeant Pharmaceuticals (VRX).

It has caught my eye that William Ackman, a long-time booster (and holder) of VRX, has joined the board of the beleaguered drug firm.  He’s also raised something like $800 million by selling shares of Mondelez, which was reportedly the largest position in his hedge fund.

To the extent that one believes in the last bull theory, and if Mr. Ackman is in fact the last bull, he’s in a no-win situation.

how high is “high” for the S&P 500 now?

One of the most reliable aspects of human behavior is that we all extrapolate from recent prior experience.  We form rules that are at first provisionary, but which gain strength as new experience seems to validate them.

This is the psychological basis for one of the few really powerful axioms of technical analysis, support and resistance.

The idea is that people buy financial assets at prices that in the past have proved profitable entry points and sell at prices that have shown themselves to be relatively high.  Put in more negative, but still psychologically valid, terms, I think, people who have previously sold at low points and bought at highs rue their decisions and reverse them if given another chance.   This provides, as it were, a built-in clientele of buyers at previous lows and sellers at prior highs.

How do we apply this insight to today’s S&P?


The S&P 500 peaked at:

2110 on February 20, 2015

2108 on March 20, 2015

2130 on May 25, 2015

2124 on June 23, 2015

2128 on July 20, 2015

2102 on August 17, 2015

2110 on November 15, 2015

2102 on December 1, 2015, and

2078 on December 29, 2015.


We closed yesterday at 1986.  Fundamentals aside, about 5% – 6% higher than that we run into a wall of people who have been trained that 2100 or so is a time to sell.

Such resistance levels aren’t insurmountable obstacles.  Breaking through them, and therefore beginning a new buy/sell training regime, would be a hugely positive sign.  But the large number of recent instances when 2100-2130 has proved a market high water mark argues that this is a formidable barrier to advance.