Trends and Backtesting

[Shanghai]

Thought I would pop in with a few questions for the wise here. I've been trading shares for many years using mainly FA only but trying to time entry and exit points with some crude TA. have been very happy with my returns, however being a curious type I though I would delve into TA a bit to see whether there is science behind it or more casting of chicken's entrails!

Some points if I may.

With regards to support and resistance, I can see the logic of horizontal lines acting as support or resistance due to the fact that this equates to where people have bought and sold. However when the lines are inclined why would they act as support or resistance at all? After all no-one may have bought or sold at those prices. Or is it more a case of people all drawing the same lines and making the same conclusions?

Anyone one (completely unrelated). I have been running a few backtests on Metastock using various trading strategies (standard Metastock plugins) just for my own interest. I am surprised that for companies that have performed well, in the majority of cases a buy and hold strategy beats most trading strategies, even with no allowance for trading costs. One area where the strategies do seem to work is on companies that have performed comparatively poorly where some trading strategies perform far better than buy and hold. It seems that the strategies I have looked at act more as effective ways of preserving capital rather than as a way to trounce the market.

As much of my FA is concentrated on small caps I would be interested as to how relevant people think many TA techniques are on small caps. Are the techniques in general equally applicable on small caps given the much lower volumes in many cases?

 

[new_trader]

With regards to support and resistance, I can see the logic of horizontal lines acting as support or resistance due to the fact that this equates to where people have bought and sold. However when the lines are inclined why would they act as support or resistance at all? After all no-one may have bought or sold at those prices. Or is it more a case of people all drawing the same lines and making the same conclusions?

I'm a little confused by this paragraph. Buy/Sell transactions have taken place along every point of a chart, not just the areas of support/resistance. In basic terms, Support is an area where there are more buyers than sellers and resistance is where there are more sellers than buyers.

 

[Shanghai]

Yes I probably didn't explain myself very well. As I understood it (and as always I am happy to be proven wrong!), one of the reasons that the support and resistance levels worked was that traders remember the price they bought or sold at, and support and resistance lines are popular places to buy and sell. If a price had fallen below a support line and then risen again, traders that had bought at the previous support line may be tempted to sell at that level to break even and hence the level becomes a resistance as more sellers dominate. But this wouldn't apply if the support and resistance lines were not horizontal.

 

[new_trader]

Yes I probably didn't explain myself very well. As I understood it (and as always I am happy to be proven wrong!), one of the reasons that the support and resistance levels worked was that traders remember the price they bought or sold at, and support and resistance lines are popular places to buy and sell.

No, not exactly. Remember, traders are buying and selling at different levels. Supp/Res is where more traders have either bought or sold. I don't understand what you mean by "worked"? I think you are putting the cart before the horse.

these articles might clear things up:

http://stockcharts.com/school/doku.php?id=chart_school:chart_analysis:support_and_resistance

http://www.investopedia.com/articles/technical/061801.asp

 

[Shanghai]

Thanks for the reply New Trader. I'll have a look at the links you posted. I can never get enough of reading.

My main point really came from the Martin Pring book I have just finished. He is talking about how once a support zone has been violated it reverses its role to resistance. He says "One rationale here is that people who bought at point B1 (resistance level) watched as the price of their stock plummeted. When it rallied back to point B2 (same level but further on) they saw their opportunity to break even, and they sold".

That's where my query about non-horizontal support and resistance comes from as the above would not then apply.

 

[TWI]

However when the lines are inclined why would they act as support or resistance at all? After all no-one may have bought or sold at those prices. Or is it more a case of people all drawing the same lines and making the same conclusions?

If the sentiment is bullish as time passes traders will be prepared to start buying at rogressively higher prices when prices dip and will ony sell at progressively higher prices as they rally. Result is uptrend. As for the reason lines can join these highs/lows. I think that is because everybody likes to quantify where to buy or sell and what simpler way than a straight line to create your trading plan. Without a plan and discipline people will not make money so I think it is self fullfilling.

 

[jasonnoguchi]

Some things you must understand about support and resistance levels are :

1. Like all technical analysis, it is NOT a precise science. A support level can become a resistance level and that "level" is not a single price but across a narrow price range.

2. Like all technical analysis, you call an apple "apple" only after you identified evidence that it is an apple, right? You cannot PREDICT or GUESS resistance and support levels. You must have the technology to see what investors are doing at these levels as evidence before you can say it is a resistance or support level.

3. It is not a magic formula. Like all technical analysis, you need to see how it works in conjunction with your overall strategy to make sense. A resistance level for a short term trader may be nothing in the eyes of a long term trader.

I see you are still just starting out in TA and somewhat eager to get into the heat. Always remember one thing, there is no magic formulas.

 

[Shanghai]

Thanks jasonnoguchi interesting points. In fact my background in fundamental analysis may actually be a hinderance in a way. With FA, understanding the relationship between cause and effect is very important, how changes in interest rates in one country affect sectors in another or how earning surprises have historically affected share prices for different companies for example. With TA, I was initially imagining that there would be a similar relationship, i.e. that given a certain set of circumstance, investors would react in a "predictable" way, and that a knowledge of this may enable traders to make a profit. Hence my question about what is actually going on in investors heads that would create a support / resistance zone for example.

I may never change the stratgy I have at the moment as it is fairly low maintenance and delivers quite acceptable profits. However I just love learning new things so I thought I would pop over to learn a bit more about TA, so thanks for all the responses.

 

[jasonnoguchi]

I must say TA is very reliable for short term trading while FA is very reliable for long term trading. I have been trading a 100% TA strategy and profiting full time for the past 11 years.

 

[alexander]

It seems that the strategies I have looked at act more as effective ways of preserving capital rather than as a way to trounce the market.

Actually, it may appear so because effectively what these poor thought strategies do is break-even at best.

You're possibly looking at the wrong strategies. Most plugin strategies are optimized filters for a certain period of time and at best their performance reverts to mean zero returns after a while. In the worst case you are getting trounced.

If you want to preserve capital you buy some triple-A bonds. It's very disappointing to trade and end up just preserving capital, I agree with you. You should look into developing your own strategies based on the experience you have gathered about the markets. Most of us have been through what you correctly discovered and end up burning a lot of midnight oil to come up with a reasonable system. By reasonable I mean a system that returns about 20% without assuming excessive risk.

Alex