Does Traditional Trading Advice Work?

[windsurfer 10]

Certainly I will concede that there is a difference between a trend and a trend trading system. My focus is on the latter, and my statement to be more precise is that the common trend trading systems (MAX, MACD, B/out) do not yield positive returns on daily equity index data.

 

I also agree that you can with optimization find positive results- simply reversing slow and fast variables in MACD will do it, but that is hardly trend-trading..

 

I'm catching a flight now to a conference (not trading) and won't be able to respond further until Monday.

[DbPhoenix 42]

  windsurfer said:

..... my statement to be more precise is that the common trend trading systems (MAX, MACD, B/out) do not yield positive returns on daily equity index data.

 

Now that I will likely agree with, the key word being "system".

 

Beginners attracted to this thread should understand that practically any idea -- whether rocket science, not rocket science, or downright harebrained -- can be made to appear profitable, whether by accident or by design, in a backtest. A more meaningful test is a properly-conducted forward test. And the most meaningful of all is a real-time test, preferably with real money.

 

Belief or the absence thereof is not pertinent. The account balance is the final arbiter.

[erierambler 10]

  DbPhoenix said:

Now that I will likely agree with, the key word being "system".

 

Beginners attracted to this thread should understand that practically any idea -- whether rocket science, not rocket science, or downright harebrained -- can be made to appear profitable, whether by accident or by design, in a backtest. A more meaningful test is a properly-conducted forward test. And the most meaningful of all is a real-time test, preferably with real money

 

Yes , the forward test proves the results and it is not easy to follow, but I disagree again with the notion " can be made to appear profitable". That is not what I do. I put in not only time but effort and it costs money as well. Because the data does not work on a daily timeframe doesn't mean the idea / strategy is not profitable. One has to find what is . One has to look. One does not make it profitable, one just has to test it. Otherwise how the hell would one know whether they are on the right track or not. I"ll leave you's alone now..........

good trading to you's

 

erie

[BlowFish 10]

  erierambler said:

 

Remember to keep an open mind and try all timeframes and different combination of ma's.

 

erie

 

Some would argue that is curve fitting. All mechanical systems are prone to stop working. Some even start again. Some would also argue it requires quite a lot of skill and discretion to keep a system running well.

 

Personally if I was going to approach something like this I'd be inclined to take something that is more based on market structure as a starting point. Maybe a Donchian channel or something....hmm that sounds familiar.

[thalestrader 14]

Here is one of my trades from yesterday, long SWN.

 

I put MACD on the chart. MACD would have had you in then out then in then out, rather than rising the trend until the market reversed.

 

Profitable for brokers, not so much for traders.

 

Best Wishes,

 

Thales

[Tams 20]

  thalestrader said:

...

I put MACD on the chart. MACD would have had you in then out then in then out, rather than rising the trend until the market reversed.

Profitable for brokers, not so much for traders.

Best Wishes,

Thales

 

 

MACD is an Oscillator.

 

An Oscillator is never meant to be used by itself.

 

An Oscillator is always used as a complement to another indicator,

e.g. a Trend indicator.

[DbPhoenix 42]

  BlowFish said:

Some would argue that is curve fitting.

 

Which is not necessarily intentional, which is why beginners so often go astray with this sort of testing.

 

If one is testing something, he has to do more than just "look". He has to do more than just "find what is". One has to do more than "just test it". Testing requires inputting data, which requires making choices. If the results are unsatisfactory, different choices are made and the testing begins again. And again. And again. Until one either quits or finds something that appears to be profitable.

 

Which is why backtesting can do no more than provide clues for further research. The sort of backtesting results posted earlier that demonstrate profit are meaningless, much less demonstrate that a moving average crossover system "works".

[thalestrader 14]

  Tams said:

MACD is an Oscillator.

 

An Oscillator is never meant to be used by itself.

 

An Oscillator is always used as a complement to another indicator,

e.g. a Trend indicator.

 

 

I beg to differ, my Good Tams. I may not use indicators any longer but I studied and tried most of the common ones and a good number of the more obscure ones. In fact, I could fund several decent small starter accounts with the money I have spent on various software add ons etc. as I made my way through the 38 steps that DbPhoenix has posted here at TL.

 

MACD is certainly presented by Gerry Appel, its developer, as a trend indicator. In fact, chapter 8 of his Technical Analysis is entitled, "Advanced Moving Average Convergence-Divergence(MACD): The Ultimate market Timing Indicator." He argues that across multiple time frames, from monthly to intraday, MACD can be applied in the analysis of market trends, "frequently capable of producing precise entry and exit signals" (Appel, p. 165).

 

He also suggests that while MACD "oscillates" above and below a zero line, that it is not best to utilize it as an overbought/oversold indicator, but that its real value is in its ability to "use divergence to recognize the most reliable signals" (Appel, p. 171). Since divergence accompanies the most reliable signals, then that in and of itself would bring into question the indicators use as an overbought/oversold indicator. Think of it this way: the signal would occur when the indicator is less, rather than more overbought/oversold, but at a price level higher/lower than the most extreme reading. In other words, using MACD as an ob/os oscillator could prove hazardous to you equity.

 

Furthermore, MACD is constructed of moving averages. Moving averages are themselves trend indicators and not indicators of overbought or oversold conditions.

 

Also, the origin of this part of the discussion in this thread was a claim that trend trading "doesn't work," i.e. it is not profitable, and the example used was a system based upon trading MACD signals.

 

So, while I have the greatest respect for you, Good Tams, I must nonetheless respectfuly disagree with your characterization of MACD as an oscilator and not a trend indicator, as well as your suggestion that MACD needs yest another iindicator to, I imagine you would say, "filter" the signals.

 

Best Wishes,

 

Thales

Edited June 12, 2009 by thalestrader
spelling

[Tams 20]

May I address you Good Thales? ;-)>

 

 

You are correct in ALL counts.

(and thank you for the references.)

 

 

MACD does show trend.

 

I call MACD an oscillator because it varies around an equilibrium point.

Calling MACD an oscillator does not mean to say that it does not show trend, nor not trend related.

 

At the same token, not all oscillator must show OB/OS.

 

 

A bounded oscillator may deem OB/OS when it is at its boundaries, when complemented with a trend indicator.

 

I would not suggest interpreting a boundless oscillator (e.g. MACD) the same way as a bound oscillator.

Edited June 12, 2009 by Tams

[jidrool 10]

Until you are making decisions from your own experience and not trite sayings you won't be able to trade profitably. How do you get to that point? Putting in your time making real time trades and taking your lump$!!!!!!! Just view it as paying your tuition. Play small until you can start to define who you are. Your way won't be my way grasshopper! One thing I can state with certainty is that you can not win at this until you conquer your emotions and you must trade real money real time for quite a while to discover how to accomplish that. Good luck and remember always lick your wounds and summon the courage to face tomorrow

[windsurfer 10]

I have a quibble with the "learn from your experience" school of trading.

 

1. Methods that work for a beginning trader in the short-term may very well not work longer term. Almost every good system has losing months. Almost every bad system has winning months. My point here is that you cannot trust short term personal experience.

 

2. There are good methods. Some have been shared. Find them. Following them may very well lead to better results than personal experience. In addition, trying to adapt them and improve them will often be a better use of time than looking at your own trades.

 

3. I favor the "scientific" school. Learn the pitfalls of testing- there are some good books and articles on this. But then test. Use multiple systems. Keep detailed records. Trade scientifically. Retire young.

[jidrool 10]

I don't disagree with anything windsurfer says. I should have qualified my remarks by saying experience comes after and during constant study and testing. There is no need to try and reinvent the wheel but we cannot underestimate the human factor in trading. I feel that our human tendencies can be our greatest impediment and until you feel and learn how to control them you will be in for a difficult time. Was it Kipling who said "if you can keep your head while all the other traders are losing theirs you will make alot of money my son.." or something like that

[jidrool 10]

In a more practical vein let me share today's rather unpleasant lessons. I have been having some success in the last several weeks by fading moves to either pivot points or the support/resistance lines that develope during the day when accompanied by a divergence in or extreme reading in the tick. Very few held and twice they broke down so quickly that the trade cost me more than I had expected. Are we entering a downtrend where this method won't be profitable for awhile, at least at support levels?Should I only take them at resistance. Just giving you a glimpse into the workings of a fledgling trader's brain? Hope it helps and if anyone has the answers to "life's most persistent questions" I'm all ears.

[DbPhoenix 42]

  jidrool said:

Was it Kipling who said "if you can keep your head while all the other traders are losing theirs you will make alot of money my son.." or something like that

 

Or, as some have suggested, "if you can keep your head whilst all the other traders are losing theirs, you have no idea what's going on."

[windsurfer 10]

  jidrool said:

In a more practical vein let me share today's rather unpleasant lessons. I have been having some success in the last several weeks by fading moves to either pivot points or the support/resistance lines that develope during the day when accompanied by a divergence in or extreme reading in the tick. Very few held and twice they broke down so quickly that the trade cost me more than I had expected. Are we entering a downtrend where this method won't be profitable for awhile, at least at support levels?Should I only take them at resistance. Just giving you a glimpse into the workings of a fledgling trader's brain? Hope it helps and if anyone has the answers to "life's most persistent questions" I'm all ears.

 

It sounds like a good system. I wouldn't try to guess long or short as long as you're not fighting the trend of the day. Good luck.

[DbPhoenix 42]

Can you post a chart that illustrates what you mean by "the support/resistance lines that develop during the day"?

[marcj 10]

Guys

 

this to big of a topic to discuss here and now. I been doing this website for years; it is my thoughts on these subject.

trading-seminar

I hope that helps

 

marcj

[rajatheroyal 10]

Hi guys i feel these tradional things dont work.coz they tell contradictorythings for eg:

1)let your porfits run.

2)you dont lose ur shirt booking profits.:doh:

how can be both true.

if these things worked then 95% traders wont be losers.

the only way is to develop a mechanical system and follow it strictly.

i dont think u can make money provided u are required to take decisions during trading hours.

[marcj 10]

rajatheroyal

 

I don't understand your statement; both can't work because i don't understand the u shirt statement.

 

but 95% of traders fail because the don't have the discipline to follow their plans or they don't have the skill to put together a good plan.

[zdo 20]

  rajatheroyal said:

Hi guys i feel these tradional things dont work.coz they tell contradictorythings for eg:

1)let your porfits run.

2)you dont lose ur shirt booking profits.:doh:

how can be both true.

if these things worked then 95% traders wont be losers.

the only way is to develop a mechanical system and follow it strictly.

i dont think u can make money provided u are required to take decisions during trading hours.

 

raja,

 

They can both be true - because market truisms are system specific (and this applies across 'discretionary', 'mechanical', and 'automated' systems btw)

 

Some systems you got to stay in or be ultimately doomed

Some systems you got to get out quickly or be ultimately doomed.

Some systems you got to do both - making each of the opposing truisms true to its proportionate degree for the particular system.

 

The spouters of truisms usually just spew them out without qualifying the type of system it applies to - omitting or not realizing that their truism is not general but instead is very system specific.

 

Most traders who never followed automation to near its limits never realize how far their exit strategies are off from optimal... ie how they are based on truisms and / or on neurological comfort zones, etc...

 

zdo

[marcj 10]

zdo

 

I think you get it. I said it like this at my website; trading-seminar.com

 

Why No Specific Plans Here?

 

Plans are all personalized. What works for me would not fit your personality. Example; I'm don't have patience, so I day trade, waiting for grass to grow and watching trades go negative for a day or two just does not work for me.

You have to understand the plan too well. You can only get that from doing it yourself.

[rajatheroyal 10]

  zdo said:

raja,

 

They can both be true - because market truisms are system specific (and this applies across 'discretionary', 'mechanical', and 'automated' systems btw)

 

Some systems you got to stay in or be ultimately doomed

Some systems you got to get out quickly or be ultimately doomed.

Some systems you got to do both - making each of the opposing truisms true to its proportionate degree for the particular system.

 

The spouters of truisms usually just spew them out without qualifying the type of system it applies to - omitting or not realizing that their truism is not general but instead is very system specific.

 

Most traders who never followed automation to near its limits never realize how far their exit strategies are off from optimal... ie how they are based on truisms and / or on neurological comfort zones, etc...

 

zdo

 

Yes sir what u said is absolutely correct when u say about systems coz anybody who has a good system will survive but many many discretionary traders dont even have a system.

They completely trade on discretion some time they book profit taking in mind " you dont lose ur shirt rule"(when time is to stick with ur positions) some times they " let their profits run(when its actually time to book the profits) so i told that it wont work for them.

For those who follow systems,they dont even have to care about these these traditonal trading advices.

for example they say never trade against the trend

then what are mean reversion systems for ?

Edited July 28, 2009 by rajatheroyal

[StockTrader6080 10]

You need to build your rules ONE AT A TIME, and then stick to them.

 

For instance for my long trades, I use rules like these:

 

Average Daily Trade Range % (14) > 5 [means the stock moves at least 5% a day]

Current Price is at least 5% < yesterday's close price [means only buy after a 5% drop]

Current Price > 20 Day Simple Moving Averge [means it is still in a long-term uptrend.

 

I simply get more winners than losers and make at least 5% on every trade.

Keep it simple and ALWAYS stick to your rules!

 

On another note, Beware of fund managers doing window dressing for the next few days as we approach the end of the month!

 

Don't get caught too Long when the market tumbles!

 

I don't believe this run-up is real for a second.

 

I've changed my automated traders to not be in more than 5 long and 5 short positions at any time.

 

Keith

 

------------------------------------------------------------------------------

Goodbye mutual funds Hello CoolTrade Robotic Traders