Technical Analysis Is Nonsense

Tsukasa Buddha

Other (please write in)
Joined
Sep 10, 2006
Messages
15,302
To prevent derail:

A typical price chart for a stock will show the price of the stock over time. A chartist looks for patterns in charts, patterns he believes are predictive. Is there any scientific evidence that there are some truly predictive patterns in stock price charts? No, but there is scientific evidence that chartists see patterns in random processes (Malkiel 2007). According to Thomas Kida, there is a wealth of scientific research going back nearly half a century demonstrating that "charting is useless." "Technical analysis can't beat the market" (Kida 2003: p. 122). The long and the short of it is: past prices of stocks are extremely poor predictors of future prices. Why are there so many technical analysts working for brokerage firms, then? Because they recommend a lot of trades and trades generate commissions (Malkiel 2007).

Does the fact that the scientific evidence shows that technical analysis is no better than chance in predicting the price of stocks mean that nobody ever makes any money buying and selling stocks following the advice of a technical analyst? Of course not. By chance alone many brokers are going to make recommendations that make money for their clients. The illusion is in thinking that technical analysis provides a magic formula that will continue to bring success forever and ever. No, in the long run, there will be some big winners and some big losers, but the average among most of the investors will be no better than what you would get if you let a monkey throw darts at a list of all the stocks listed on the New York Stock Exchange.

Linky.

Technical anlysis is doomed to fail by the statistical fact that stock prices are nearly random; the market's patterns from the past provide no clue about its future. Not suprisingly, studies conducted by academicians at universities like MIT, Chicago, and Stanford dating as far back as the 1960s have found that the technical theories do not beat the market, especially after deducting transaction fees. It is amazing that technical analysis still exists on Wall Street. One cynical view is that technicians generate higher commissions for brokers because they recommend frequent movement in and out of the market.

Linky.

1. Economists’ predictions are no better than guesses

Forecasting skill of economists is no better than guessimates by Main Street investors.

2. Government economists often worse than guesses

Sherden discovered that predictions made by the elite economists on the President’s Council of Economic Advisors, the Federal Reserve Board, and even the non-partisan Congressional Budget Office were actually worse than guessing.

3. Long-term accuracy is impossible

The accuracy of forecasting declines the longer the lead times.

4. Turning points cannot be predicted

Economists cannot predict the crucial turning points in the economy, confirming Siegel’s research. Worse, the vast majority of all long-term predictions fail.

5. No specific forecasters are better than the rest of pack

Sherden also learned that no particular forecasters were consistently more accurate.

6. No forecaster was more expert with specific statistics

No forecaster has consistently higher skills in predicting any one economic statistic.

7. No one ideological orientation was better

No ideology perspective consistently produced superior forecasts.

8. Consensus forecasts do not improve accuracy

But still, the press and their readers love those lists, averages and consensus forecasts.

9. Psychological bias distorts forecasters and their forecasts

Some economists are naturally optimistic and bullish. Others are naturally pessimistic bears. Some are conservative, some progressive. Why? Look inside their brains at their DNA, Every economist has mental biases and political ideologies that distort their choice of research topics and data selection, and therefore, skew their predictions.

10. Increased sophistication does not improve accuracy

Sorry folks, but all the new scientific methods, technologies, algorithms and computer models of the economy can make forecasts worse. At least give skilled Wall Street insiders an even better edge over naive retail investors.

11. No improvement over the years

Finally, Sherden says there’s no evidence that economic forecasting has improved in recent decades, despite vast new technologies.

In recent years the science of forecasting has been deteriorating more as partisan politics intensifies, along with global macro trends, high-frequency trading, Wall Street’s market manipulations, and unpredictable black swan events. All increase volatility, uncertainty and create countless new ways for forecasters to invent new illusions of economic accuracy, while hardening the mental biases of forecasters.

Linky.

This paper provides evidence that “illusory correlations”—a well-documented source of cognitive bias—lead some agents to be imperfectly rational noise traders. We focus on the head-and-shoulders chart pattern, considered by technical analysts to provide one of the most reliable trading signals. Our findings indicate that the pattern is associated with a substantial rise in trading volume even though it does not profitably predict directional movements. We further substantiate the connection between head-and-shoulders trading and imperfectly rational noise trading by showing that the pattern is associated with lower bid-ask spreads.

Linky.

We conduct an extensive examination of the profitability of technical analysis in ten emerging foreign exchange markets. Studying 25,988 trading strategies for emerging foreign exchange markets, we find that the best rules can sometimes generate an annual mean excess return of more than 30%. Based on standard tests, we find hundreds to thousands of seemingly significant profitable strategies. However, almost all of these profits vanish once the data snooping bias is taken into account. Overall, we show that the profitability of technical analysis is illusory.

Linky.

The primary aim of this study is to investigate the validity and predictability of technical analysis in eight Asian equity markets. We employ the bootstrap tests of White (2000) and Hansen (2005) to determine whether any superior trading rule is found to exist amongst the ‘universe’ of technical trading rules identified by Sullivan et al. (1999). We use these powerful bootstrap tests to ascertain the profitability of technical analysis, along with two institutional adjustments for non-synchronous trading and transaction costs. The empirical results indicate that these three elements, data snooping, non-synchronous trading and transaction costs, have significant impact on the overall performance of technical analysis; indeed, the results for these eight Asian stock markets support the efficient market hypothesis, demonstrating that the generation of economic profits through the use of technical analysis is extremely unlikely with these particular markets.

Linky.

The main purpose of this paper is to investigate the validity and predictability of technical analysis in the Taiwan stock market. Bootstrapped tests of White (2000) and of Hansen (2005) are employed to ascertain whether there exists a superior trading rule among two broadly used sets of technical analysis. One coming from Brock et al. (1992) and the other from Sullivan et al. (1999). Moreover, this study brings together powerful bootstrapped tests along with two institutional adjustments to ascertain the efficacy of technical analysis: (1) non-synchronous trading and (2) transaction costs. The empirical results indicate that this triad-data snooping, non-synchronous trading and transaction costs, has a great impact on the performance of technical analysis. In fact, the Taiwan stock market stands for market efficiency, and economical profits cannot be rendered from technical analysis in this market.

Linky.

► We conduct an intraday technical analysis of individual stocks on the Nikkei 225. ► We investigate technical rules constructed by the information in the limit order book. ► We also investigate technical rules constructed by past price information. ► Past prices do not contribute to profiting in intraday trading. ► Non-execution and picking-off risks are too large to make profits in intraday trading.

Linky.

As opposed to fundamental analysis, financial chartism (or technical analysis) is based on the assumption that future movements of a traded asset can be inferred from past regularities. In spite, or maybe by virtue of its lack of scientific grounds, it enjoys an ever-growing popularity among amateur traders, as countless websites provide free, simple and powerful tools for making professional-looking charts and even automate the detection of trends. Though technical analysts account for the validity of chart patterns in terms of market psychology, constructivist explanations also abound among trading communities: it is all as if economic agents reflexively used technical analysis as a means of coordination on the market. Markers like moving averages and trend lines serve as self-referent indicators whose efficacy only gets reinforced as more agents use it as pivotal points. Performativity is the technical analyst’s most natural language: explicit mentions to mechanisms such as the self-fulfilling prophecy or the Keynesian beauty contest are the only justifications s/he needs. Some chartists even claim their discipline does not contradict the efficient-market hypothesis: if prices already reflect all past publicly available information, one may skip the hassle of dissecting balance sheets or mastering pricing models, focusing instead on candlestick or diagram patterns which, by conjecture, contain all the necessary information to explain price movements. Drawing on the comparative analysis of nonprofessional trading blogs, our paper examines how these major epistemic modes of justification of technical analysis have come to support each other, in a context of proliferation of charting tools.

Linky.
 
If it worked, then people would be taking huge advantage of it, which would cause it not to work. The only methods that can work for long are ones that are not widely known, or that require large amounts of money and/or specific technology, access, and low transaction costs that are not available to the majority (automated trading).
 
The only ones that work are manipulative ones that are supposed to be illegal, unless you're really rich or something of that sort.

OR unless you own a media network and can play with everyone's fate.
 
Taleb's books are listed in the skepdic link, he has made a lot of money based on the analysis that you cannot predict the marktet. He has technical papers to back up what he is saying, but the math is beyond me at this point.
 
Just to be argumentative, shouldn't the title be "Chart Reading is Nonsense"?

It should also be pointed out that just because you can't pick tomorrow's stock prices doesn't mean that you can't tell a dependable stock from an undependable one based on the company's history (which is not based on stock price alone).
 
Just to be argumentative, shouldn't the title be "Chart Reading is Nonsense"?

It should also be pointed out that just because you can't pick tomorrow's stock prices doesn't mean that you can't tell a dependable stock from an undependable one based on the company's history (which is not based on stock price alone).

I think you're approaching "value investing" with that description, which is not techinical analysis.
 
I think you're approaching "value investing" with that description, which is not techinical analysis.

"Why would anyone do anything other than value investing?" - W. Buffett

I always thought technical analysis and chart reading were the same thing. But either way, yeah, it's nonsense.
 
The only element of techncial analysis that I think has merit is the principle of psychological breakeven that can generate upward sell flurries in the same pricepoints we saw downward buys earlier.

I take the last year's chart into consideration when choosing an entrypoint for what is otherwise value investing on any stock I individually select (95% of my portfolio is a balanced fund with the equities portion based on an index).

eg: AAPL $530.
 
I take the last year's chart into consideration when choosing an entrypoint for what is otherwise value investing on any stock I individually select...

Wouldn't it make more sense to determine your entry point based on the value of the business, rather than some ticker price?
 
The only element of techncial analysis that I think has merit is the principle of psychological breakeven that can generate upward sell flurries in the same pricepoints we saw downward buys earlier.

I agree. I think there is some physiology that could be exploited. Increasingly computers are better at doing that exploiting than human short term traders. Ultimately it’s a zero sum game where most people are at a disadvantage.
 
Interesting article in Wikipedia about Bollinger Bands (a form of technical analysis) which references some studies of its effectiveness. It seems that it was actually more effective to do the opposite of what the method recommended. Only two studies, though...too small to draw solid conclusions.
 
Wouldn't it make more sense to determine your entry point based on the value of the business, rather than some ticker price?

Generally the entry point is "I have some cash to invest" so once a value stock has been identified, the question I'm addressing with this analysis is whether it might drop temporarily within the investment timeframe and could there be some value in a short delay.

This is a small tweak on 'market timing' which is another controversial strategy.

The benefits are slight and there is risk that inertia could be so strong that it will push through a trading range very quickly, so if somebody preferred to keep it simple and dollar-cost-average I think that's very sensible too.
 
Generally the entry point is "I have some cash to invest" so once a value stock has been identified, the question I'm addressing with this analysis is whether it might drop temporarily within the investment timeframe and could there be some value in a short delay.

Just out of curiosity, how do you determine this? Do you have some sort of formula?
 
Just out of curiosity, how do you determine this? Do you have some sort of formula?

Not a formula. What I do is look at the prior year's trading. Sometimes a value stock has lost a great deal of market cap, so if I'm buying it I'm predicting a recovery over time.

What's worth observing is whether it's been bought in large volumes slightly above this pricepoint on the way down within the last 24mos. If so, it may decline awhile as people sell their positions 'to break even'.

So, for example, I was very comfortable buying AAPL at $390 in Summer because the decline was after very small volumes. Later, when I considered a purchase at $525, I felt it was going to rise to $530 and drop down b/c tens of millions of shares changed hands at $530 within the last year and there's so much 'hanging overhead'. This turned out to be the case, and two weeks of delay allowed me to buy at $512, after which it has climbed back up through $530 without resistance, as the $530 breakeven sellers were depleted.

In this sense it's not very technical, but I do use 'past history' to anticipate future events. But I'm not extrapolating patterns or trajectories.

The underlying assumption is that most retail investors hang onto a stock through decline and recovery until it returns to exactly the original purchase price, so they don't realize 'dollar' losses. (even if the less tangible opportunity to purchase a better stock in the meantime has been lost instead).


ETA: a high-volume investor may also use this theory to use cash elsewhere temporarily, expecting a preferred equity to be trading in a range until the hanging overhead is depleted - they know the opportunity will park for a week or two and the ROI for that timeframe is zero, so capital should be deployed elsewhere.
 
Last edited:
The underlying assumption is that most retail investors hang onto a stock through decline and recovery until it returns to exactly the original purchase price, so they don't realize 'dollar' losses.

That may be true more times than not, but in terms of average potential profit it is probably balanced by capitulation.
 
That may be true more times than not, but in terms of average potential profit it is probably balanced by capitulation.

Keynes noted that "The market can stay irrational longer than a person can stay solvent", so yeah, sooner or later capitulation has to come into play.

On the OT, I don't claim that charts can micro-time the market (as in "Sell GE no later than 10:35 on Monday, August 8th"), but charts CAN help observant investors spot broad trends and potential problem.

Two examples:

1) Bubbles: Every major investment bubble, from Tulip Bulb Mania to the Housing Meltdown followed basically the same pattern, so when you see that bubble pattern starting to develop in ANY market or the market as a whole, it's time to be ready to run.

2) Strong statistical correlations, such as the "Hindenberg Omen" with well documented histories of accuracy should, at a minimum, put any prudent investor on watch

http://wallstcheatsheet.com/stocks/this-stock-market-omen-has-been-confirmed.html/?a=viewall

On a related note, if ANYone tells you his theory has any origins in Efficient Market Hypothesis...run...don't walk, RUN. EMH is right up there with Flat Earth and Spontaneous Generation in terms of BS theories.
 
Tulip Blub Mania...

Side note, there is a lot of debate if there was really was a major tulip mania, it seems to have been confined to a very small group of people, with very limited economic fallout.

http://en.wikipedia.org/wiki/Tulip_mania#Modern_views

Yes, some of the critics are proponents of, Efficient-market hypothesis, but that doesn't mean the their analysis of the history was wrong. Most of the mania myth was based on one story by Mackay.

Let's just say there's a lot of room for doubt.
 
Last edited:
1) Bubbles: Every major investment bubble, from Tulip Bulb Mania to the Housing Meltdown followed basically the same pattern, so when you see that bubble pattern starting to develop in ANY market or the market as a whole, it's time to be ready to run.

Actually, it's time to accumulate cash, so you can buy when there is blood in the streets.

2) Strong statistical correlations, such as the "Hindenberg Omen" with well documented histories of accuracy should, at a minimum, put any prudent investor on watch

http://wallstcheatsheet.com/stocks/this-stock-market-omen-has-been-confirmed.html/?a=viewall

:dl:
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom