Did you suffer before doing well or vice versa?

[theasxgorilla]

.
Until I started making some really stupid moves, some panic selling and buying and with the maket so turbulent my reading seems to be off non stop. Accounts down 30% on principle and I am starting to wonder.

I assume by 30% down on principal you mean on the initial amount you started with, before doubling your money. If this is the case then a rough back of the napkin calculation says you have experienced a 65% peak to trough portfolio equity drawdown. Hmmm.

Get out and stay out, until you can put into perspective where you (WE) are within the market cycle and what kind of market we're currently in. IMO, the writing has been on a wall for a little while now already...noisy (volatile) and trending (down).

There are other threads which talk about being in the top 3%/5% of traders who are profitable. Quiet (not volatile), trending (UP) markets can make nearly all of us part of that 3%/5%...go figure.

Also, panic=emotion driven-decision making. Emotion in trading is OK (IMO). Many experts will tell you it's not, but for most people it's normal to need to work to silence the voices within. Emotion driven decision-making is a recipe for bigger drawdowns than necessary.

Step back, get perspective. Validate your plan or make a plan before you commit to executing it. Then, when the time is right, execute the plan, manage it, and spend the lion-share of your efforts on psychology.

As for details on plans and psychology, use the forum. There a literally dozens of experienced traders here who are only too willing to help others understand how they are succeeding.

ASX.G

 

[MichaelD]

KNOW = know within the limits of reasonable due process.

The key hallmark of a profitable trading strategy is that you KNOW in advance what the strategy's performance parameters are.

You KNOW what the maximum expected drawdown is.

You KNOW how it performs in various market conditions.

You KNOW its expected run of losses.

In short, you KNOW in advance when the strategy will be declared not to be working any longer and what to do if this happens.


If these parameters are exceeded, you STOP trading the strategy and evaluate what is going wrong;
- is the strategy no longer valid
- are you trading the strategy incorrectly

Once you have worked out what is going wrong and fixed it then you start again.

All else is just gambling.

Most likely, your strategy is a bull market only excessive risk strategy which has now been slaughtered by the bear market. (AKA gambling and having a run of luck which has now run out).

 

[chops_a_must]

KNOW = know within the limits of reasonable due process.

The key hallmark of a profitable trading strategy is that you KNOW in advance what the strategy's performance parameters are.

You KNOW what the maximum expected drawdown is.

You KNOW how it performs in various market conditions.

You KNOW its expected run of losses.

In short, you KNOW in advance when the strategy will be declared not to be working any longer and what to do if this happens.

See, I don't think anyone can ever KNOW (in your terms) a lot of these things - unless you believe that markets are inherently predictable by way of analysing historical data, in which case, every system once profitable will always remain profitable.

I think a lot of people here, especially the mechanical traders, would do well to look into the history of the word, "expectation".

Having said that, it doesn't take all these fancy terms to know if something is not working. Long only at the moment, will not work, and a 30% peak to trough drawdown on a very short term system, to me, is not working.

P.S. - what would you do with a trading strategy Michael, that has a theoretical statistical/ probability edge, but cannot be backtested adequately?

 

[theasxgorilla]

See, I don't think anyone can ever KNOW (in your terms) a lot of these things - unless you believe that markets are inherently predictable by way of analysing historical data, in which case, every system once profitable will always remain profitable.

I knew someone would mention this

I was going to put a bracketed caveat in my post about knowing what kind of market we are in, and where we are in the cycle. The reality/truth/fact! is that we can really only know where we've been. We can only make a true statement which says, 'up until now and based on my analysis, since the peak in Nov 07, we have been in a noisy and down trending market'.

Of course, the next days trading might be the beginning of a different kind of market. My opinion is that no-one can predict.

ASX.G

 

[sidswingerhead]

Hi Tech/A,

I agree that having a plan doesn't automatically guarantee your profits. But it will assist you in the most important thing, that is to protect your capital and minimise your risk. Not having a plan at all will surely lead you to your doom.

I also agree that the blue print as you say is ever evolving depending on circumstances influencing the market at the time.

How do you design or modify a profitable plan without using it?

Paper trading is IMO not beneficial or truly practical because the moment you put real money into the market your mental bias takes over. Paper trading has no emotion.


I disagree with you vehemently disagreeing about abiding by your rules

What is the point of setting a plan or building rules, if continually neglect to act upon them? True improve or modify them as situations change.

I would appreciate your input into what you consider makes a good blue print.

 

[tech/a]

Sid

I agree that having a plan doesn't automatically guarantee your profits. But it will assist you in the most important thing, that is to protect your capital and minimise your risk. Not having a plan at all will surely lead you to your doom.

Provided that components of the plan actually do minimise risk.
Provided that the person designing the plan actually understands how best to apply risk mitigation. So in essence I agree.

How do you design or modify a profitable plan without using it?

You can certainly forward test by using it as you say but ultimately many I have found that this method is/can be very costly not to mention some cases emotionally draining.
There are very good software around for testing ideas,Amibroker/Metastock/Tradesim to name a few.Plus some very talented people who can code for you if your not up to speed. For forward testing ideas without code but you need lots of patience and time STATOR.

What is the point of setting a plan or building rules, if continually neglect to act upon them?

The point is you don't know their profitable most don't know why they could be profitable or doomed to failure.
Very few very talented people can design and trade methods which cannot be tested purely because they understand the building blocks required for a successful trading methodology/plan.

True improve or modify them as situations change.

This again I have to disagree with strongly.This is the downfall of many a trader and their plan. Constant tinkering/twiddling with any methodology without first knowing ahead what this fiddling does to the performance of your plan,has many traders totally confusing and disappointing themselves.

If the KNEW that by altering their exit it would decrease strings of losses,increase drawdown but increase number of winning trades---then they could tell if when applied in the REAL world whether the change was working as tested.They have something to benchmark.

I would appreciate your input into what you consider makes a good blue print.

We all have different ideas on what makes a "good blueprint".
My definition.(In this market)

Positive expectancy.
Smoothness of equity curve.
Frequency of trades
High win rate
Short trade length
Reward to risk above 1.5:1

ASX

I was going to put a bracketed caveat in my post about knowing what kind of market we are in, and where we are in the cycle. The reality/truth/fact! is that we can really only know where we've been. We can only make a true statement which says, 'up until now and based on my analysis, since the peak in Nov 07, we have been in a noisy and down trending market'.

Of course, the next days trading might be the beginning of a different kind of market. My opinion is that no-one can predict.

But we can anticipate.Many here have for a long time indicated 4700 ish. Many here are no longer trading long term long methodologies. Although we did for years. Many are doing very little and many are trading to the choppiness seen in this market and will I anticipate be with us for quite a few years. As for the next day---anticipation is for a rally towards 6000 now for a while.So shorter term long positions would be favored.

Chops

See, I don't think anyone can ever KNOW (in your terms) a lot of these things - unless you believe that markets are inherently predictable by way of analysing historical data, in which case, every system once profitable will always remain profitable.

I agree in part to your points.While the market may not be finitely predictable the way a trading methodology is put together will determine how predictable its performance is/can be. It will vary in performance between the mean and the high and low of performance due to those unpredictable movements in the market. Any trading methodology can only be expected to remain profitable IF it is traded in conditions in which its been designed.
If those conditions change enough then it will cease to work.

Having said that, it doesn't take all these fancy terms to know if something is not working. Long only at the moment, will not work, and a 30% peak to trough draw-down on a very short term system, to me, is not working.

You could be right but there also could be exceptions to your qualifications of "Not working" they may still be in their Blueprint and quite happy with their results.

I think a lot of people here, especially the mechanical traders, would do well to look into the history of the word, "expectation".

True,and some/most traders would upon failure and study into that failure realise that they had Unrealistic expectations of their plan.

Chops

Just on mechanical trading.
Profitable mechanical trading systems don't come about on your first plug-in of conditions and variables. I'm sure all those who have walked the mechanical path will tell you that you soon learn what combination of Conditions and Variables DONT WORK..
You also learn what happens to a trading methodology when you play around with stops/trailing stops and position sizing. Important aspects of plans which are rarely considered by the masses.

Having a grounding in mechanical testing capability would in my view place you way way ahead of the crowd in developing a discretionary trading method (perhaps one that cannot be coded) which can have an EXPECTATION of profit well beyond a method designed by someone who hasn't.

 

[sidswingerhead]

Sid

Provided that components of the plan actually do minimise risk.
Provided that the person designing the plan actually understands how best to apply risk mitigation. So in essence I agree.


You can certainly forward test by using it as you say but ultimately many I have found that this method is/can be very costly not to mention some cases emotionally draining.
There are very good software around for testing ideas,Amibroker/Metastock/Tradesim to name a few.Plus some very talented people who can code for you if your not up to speed. For forward testing ideas without code but you need lots of patience and time STATOR.

Forgive me but haven't you gone from developing a trading plan to developing a software based trading system?

(IMO) The 2 are totally separate, a plan or set of rules by which you employ a strategy for risk management, entry and exit guidelines for minimum loss and maximum profit with minimal risk.

Building functions utilising charting software is much further advanced than having a set of rules to apply your basic trading principles.

The topic of system building and back testing is another story unto itself.

 

[sidswingerhead]

A couple of simple questions to ask yourself if you start have more loosing trades than winners.

Where did it go wrong? At the start or the end?

Was the entry strategy chosen by you correct?
If not, then what were the mistakes made? If you don't understand why your entry didn't work you will continue to make the same mistakes.


Do you continually find that your initial stop loss is being fired, only to see the SP rise shortly afterward? Do you fully understand the principles of stop losses? If no, you need to review your stop loss strategy. Research stop loss es and employ one that best suits your needs.

How often do you see your winning trade go South? If it a regular occurrence the you have "no exit strategy". You have been blinded by greed and suffer delusional incompetence, or is it because did not stick to your original plan?


It all leads to building a strong plan.

 

[ShareIt]

Hey Blaster,

Been in EXACTLY the same situation as you... was making a killing in the bull market and then all this down turn came about... sent me broke to be honest... The truth is, a bull market is a lot easier to make money in... where as now it is much harder. After I had lost all my cash, i closed my trading account and just forgot about trading for a while. I cleared all the emotional thoughts and slowly came around to that a trading plan needs to adjust to the different market conditions... to cut this story short, I have refined my plan and am making some good profits now... my advice would be to take a step back, clear the emotion and invest your money in more education (primarily books).... at the moment it sounds like you are taking semi-guesses as which way to trade... if this is not the case, then your plan is just not working and it probably need more education around it... good luck

 

[chops_a_must]

ASX



But we can anticipate.Many here have for a long time indicated 4700 ish. Many here are no longer trading long term long methodologies. Although we did for years. Many are doing very little and many are trading to the choppiness seen in this market and will I anticipate be with us for quite a few years. As for the next day---anticipation is for a rally towards 6000 now for a while.So shorter term long positions would be favored.

Who says you need to be a mech analyst to anticipate?

Chops

I agree in part to your points.While the market may not be finitely predictable the way a trading methodology is put together will determine how predictable its performance is/can be. It will vary in performance between the mean and the high and low of performance due to those unpredictable movements in the market. Any trading methodology can only be expected to remain profitable IF it is traded in conditions in which its been designed.
If those conditions change enough then it will cease to work.

But you don't necessarily know what conditions are present. If you have a system designed long only, for a trending market, it may not work if the subset of conditions change within the larger market conditions. It's why I could never be a fully mechanical trader, because I just cannot agree with the underlying philosophy behind it.

Take Niz for example. No offence to him, but it wouldn't have exactly taken Einstein to work out a long only long term trend following system would be shellacked in this market, especially since December. So my point would be you need a manual override (in my systems anyway) to stop you from defying COMMON SENSE in trading.

True,and some/most traders would upon failure and study into that failure realise that they had Unrealistic expectations of their plan.

Got nothing to do with that. It's much simpler. Just involves another word that people think they eliminate by doing all this testing.

Chops

Having a grounding in mechanical testing capability would in my view place you way way ahead of the crowd in developing a discretionary trading method (perhaps one that cannot be coded) which can have an EXPECTATION of profit well beyond a method designed by someone who hasn't.

I don't disagree with that. My response would be to find out statistics on event criteria, or code something that induces a high frequency entry around said event criteria, to replicate an essentially "non-consistent" or inability to be consistent entry. At least that way you have some kind of "valid" belief (whoops, there's a clue) or at least justification for trading in that manner.

I don't believe there is much difference in trading based on a theoretical statistical advantage, as compared with a mechanical system. Because to me, trading based on a mechanical system, is essentially a more complicated way of establishing a statistical advantage.

Cheers.

 

[tech/a]

Who says you need to be a mech analyst to anticipate?

no one ASX I thought was referring to traders in general.So am I

But you don't necessarily know what conditions are present. If you have a system designed long only, for a trending market, it may not work if the subset of conditions change within the larger market conditions. It's why I could never be a fully mechanical trader, because I just cannot agree with the underlying philosophy behind it.

This occurs from time to time.
Provided the system trades within the blueprint you can be pretty confident that market conditions haven't moved out of those used during design.When they do you can bet that conditions have changed to a significance which renders the system in new territory. Most systems exit or stop triggering trades in these times.

Take Niz for example. No offence to him, but it wouldn't have exactly taken Einstein to work out a long only long term trend following system would be shellacked in this market, especially since December. So my point would be you need a manual override (in my systems anyway) to stop you from defying COMMON SENSE in trading.

I have no argument with this,however Id remove commonsense as an indicator and replace it with sound analysis.---leading to a common sense decision perhaps. I have to say this as Ive been out of long term long trading since last July! We have kept T/T going as an evaluation process to see if it gets to a point of Failure V Blueprint---even if it does and the system stops it will show a very healthy return over the period traded.

Got nothing to do with that. It's much simpler. Just involves another word that people think they eliminate by doing all this testing.

Got everything to do with it.
If you are trading a flawed method/Plan you have an unrealistic expectation of it--to make a profit.
The thing is you don't know it!

I don't disagree with that. My response would be to find out statistics on event criteria, or code something that induces a high frequency entry around said event criteria, to replicate an essentially "non-consistent" or inability to be consistent entry. At least that way you have some kind of "valid" belief (whoops, there's a clue) or at least justification for trading in that manner.

I don't believe there is much difference in trading based on a theoretical statistical advantage, as compared with a mechanical system. Because to me, trading based on a mechanical system, is essentially a more complicated way of establishing a statistical advantage.

Cheers.

Ok the you need to know when all the statistical criteria is placed within a plan how it performs overtime.
A singular set of statistics while true in itself may not when combined with others deliver a single positive result.Logic says they should however practice proves to us often that it doesn't.

I'm not saying you cant trade any other way than Systematically/Mechanically.

What I am saying is that you need to know WHY your plan has a chance to profit.
No good writing out all these logical conditions into a trading plan ---following it religiously for 12 mths and losing $20,000 and WORSE not knowing why!

Or having a perfectly sound plan which suffers a NORMAL draw-down only to be completely ruined with constant tweaking because your scared to death your plan doesn't work.

Other than trading constantly until failure or Riches the only way I know to fast track this learning experience is to become proficient in systems testing,even if you have no intention of trading a system and end up trading in a Discretionary manner.

 

[blaster]

Wasn't expecting any more action on this one so thanks once again for replies here, there's some good stuff .

Granted - maybe I should word things better pretty it all up but - been published 8 times believe it or not , true story . But to me we're at the pup.
When you grow up in a family of 6 teachers and one genius , no not me my dad - you tend to have an incurable case of interllectual over load I'm afraid and the odd crossed t or dotted i just don't do it for you.

Or- Take our genuises in Cambera for instance they all write very nicely I'm sure. But we have me to thank for the the highest birth rate in thirty years and the baby bonus. Amongst all those brains - cough , no one could figure out what to do with the aging populatiion , or why we even had one so it is not everything , far from it.

I don't think I needed to clarify more at all it seemed clear enough to me but , then maybe not.You don't wanna put everyone to sleep do you.

The point was, yes my stratergies work very well or has for around close 12 mths now and are still for now anyway. The problem right now is me. Mind you she is one confusing time if your trying to trade that's for sure .
Writers block , or artists , which I use to be and have experienced first hand. Zilch for mths on end is my problem , traders block ! Which along with some help mentioned , is what's caused the sudden dwindling account.
Can't think , I'm doing stupid things all of a sudden .
I was wondering if traders find that they have turned corners at times making stupid moves for periods on end all of a sudden against their knowledge , better judgement - writers block. No matter how well they have done prior and does it return for them , how do they get it back ?

As an artist I know how to - by walking away . When it's ready and has something to say it returns and it's full steam ahead again . Dunno if that works for a trader though or if it's a natural accurence with and it's getting expensive

Cheers

 

[Julia]

But we have me to thank for the the highest birth rate in thirty years and the baby bonus. Amongst all those brains - cough , no one could figure out what to do with the aging populatiion , or why we even had one so it is not everything , far from it.

Since you've diverted from the topic, I can't resist asking exactly how you are responsible for the highest birth rate in 30 years (will desist from making snide comments about the obvious!), and also exactly how you were responsible for the baby bonus (probably the silliest bit of government policy ever), and lastly, I'm fascinated that you have figured out how to manage the ageing population. Please tell us.

 

[MRC & Co]

Or- Take our genuises in Cambera for instance they all write very nicely I'm sure.

Im by no means a genius, but thanks for the wrap!

Good luck with the kids! :jerry

In all seriousness, I still see no sound trading methodology explained, so its really hard to help. But all the best!

 

[chops_a_must]

Got everything to do with it.
If you are trading a flawed method/Plan you have an unrealistic expectation of it--to make a profit.
The thing is you don't know it!

Ahhh... I'm talking of the etymology of the word expectation. Nothing more, nothing less.

Ok the you need to know when all the statistical criteria is placed within a plan how it performs overtime.
A singular set of statistics while true in itself may not when combined with others deliver a single positive result.Logic says they should however practice proves to us often that it doesn't.

I'm not saying you cant trade any other way than Systematically/Mechanically.

What I am saying is that you need to know WHY your plan has a chance to profit.
No good writing out all these logical conditions into a trading plan ---following it religiously for 12 mths and losing $20,000 and WORSE not knowing why!

Or having a perfectly sound plan which suffers a NORMAL draw-down only to be completely ruined with constant tweaking because your scared to death your plan doesn't work.

Other than trading constantly until failure or Riches the only way I know to fast track this learning experience is to become proficient in systems testing,even if you have no intention of trading a system and end up trading in a Discretionary manner.

I don't disagree with this. Like I was saying above, mechanical testing and trading, for me, is a complex way of developing a statistical/ probable/ expectational bias for trading. But it by no means assures you of those results (hence my point about the origin of the word expectation). If a system requires exact entries, and your entries suck donkey balls, it's not going to work. You have to have a method that suits your own style. And yes, the combination of probabilities and your own style has to be the justification for the plan.

For me, currently, it is a matter of working towards finding various tradeable criteria. And in a high frequency manner, so it doesn't matter if they are traded in a discretionary manner or not.

But until then, it will remain simple. Continue to trade breakdowns in this market, and breakouts if we go flat to up. And gaps throughout.

Cheers.

 

[theasxgorilla]

We have kept T/T going as an evaluation process to see if it gets to a point of Failure V Blueprint---even if it does and the system stops it will show a very healthy return over the period traded.

Just on this point. I think a very large part of the reason why it hasn't traded outside of the blueprint has been the discretionary entries. If you trade TT through various out-of-sample periods using data from the last 12 or so months, taking every trade, you would rather quickly find it outside of any blueprint I've seen mentioned. Drive TT through some of the rapid correction periods we've had, since and including the June 06 correction and you'll find many of the results catastrophic. Even with just 50% margin, not the 70% that the test case is claimed to have been driven at.

That might have been useful for someone thinking about starting a long only trendfollowing system to know. But in fairness the code is public and this gives us all an opportunity to test and discover for ourselves.

Ideas, as with products and services, are a definite case of buyer beware.

ASX.G

 

[blaster]

Since you've diverted from the topic, I can't resist asking exactly how you are responsible for the highest birth rate in 30 years (will desist from making snide comments about the obvious!), and also exactly how you were responsible for the baby bonus (probably the silliest bit of government policy ever), and lastly, I'm fascinated that you have figured out how to manage the ageing population. Please tell us.

Yeah I know - Jesus talk about get carried away , tried to delete the lot later but oh no, the damn thing wouldn't delete .
Forums can be so frustrating , just trying to prove a point.

So you really think the baby bonus was a stupid idea ? What else are you going to do when everything in such a young country is so anti family no-ones having kids any more ?
Cost you allot more looking after millions of oldies later than a baby bonus now , do the sums and hey it got us breeding again !

Remember when all of our great leaders were just so baffled at what to do with a self inflicted aging population back when Latham was running.
I just said to Latham look what the home grant did for housing , it's still going , bring out a baby bonus they'll start breeding like rabits and he did.

Costello tries to take the credit for it but that was originally Lathams policy that I suggested , Howard just matched it to win .
Does seem rediculous paying people to have kids but hey when they do everything they possibly can to wipe family out then complain about an aging population - stuff em.
Now thE topic's really gone West- sorry about that.

Cheers

 

[MichaelD]

See, I don't think anyone can ever KNOW (in your terms) a lot of these things - unless you believe that markets are inherently predictable by way of analysing historical data, in which case, every system once profitable will always remain profitable.

I cannot figure out how you went from my statement of
"you KNOW in advance when the strategy will be declared not to be working any longer"
to
"unless you believe that markets are inherently predictable by way of analysing historical data, in which case, every system once profitable will always remain profitable"

Such a belief system would be illogical, irrational and ultimately lead to blowing up. All backtesting does is show you what worked in the past and give you some idea of how it worked in the past. As such, it is an invaluable tool - if it has never worked in the past, then it is not likely to suddenly start working tomorrow. It is also able to provide you some indication of when you should say "enough - this system is not performing to expectation".

Having said that, it doesn't take all these fancy terms to know if something is not working. Long only at the moment, will not work, and a 30% peak to trough drawdown on a very short term system, to me, is not working.

These statements are entirely subjective;
- there is overwhelming compelling evidence that for long term trend following it is best to stay in the market at all times, waiting out the drawdown, rather than attempt to time the market. A blanket statement that "Long only at the moment will not work" takes only a short term view and also attempts to predict the market.

- 30% peak-to-trough drawdown would be no problem at all for some traders.

P.S. - what would you do with a trading strategy Michael, that has a theoretical statistical/ probability edge, but cannot be backtested adequately?

If I could not backtest/walk forward test a strategy to my satisfaction, I would not trade it.

However, I have not yet come across a strategy that cannot be backtested and walk-forward tested adequately - it may be very time consuming and labour intensive to examine some edges, but they can be tested if you really want to.

Would you like to offer an example of a strategy that could not be backtested adequately?

 

[prawn_86]

As happens with so many other threads, it appears this has turned into a mech vs non-mech traders.

Why does it matter??

Some people dont backtest, it does not mean they are not successful or that their system does not work.

I would argue that fundamental analysis cannot be backtested, or can only be done so through research, not a mechanical system.

I think people need to realise there is more than one way to invest in the market, and as a generalisation it seems funny how many mech traders think that their way is the one and only and that everyone should develop a mechanical system.

 

[tech/a]

As happens with so many other threads, it appears this has turned into a mech vs non-mech traders.

Why does it matter??

Not at all-- you've missed the point this thread has become about validation of a trading method/plan.

Some people dont backtest, it does not mean they are not successful or that their system does not work.

Absolutely correct,what it does mean is they wont know that their plan is profitable until sometime down the track.They are trading ideas not a proven plan with all the hallmarks of a plan that can be profitable.

I would argue that fundamental analysis cannot be backtested, or can only be done so through research, not a mechanical system.

Well we have had a few try to prove this exact point,the last Ducati

https://www.aussiestockforums.com/forums/showthread.php?t=2829&highlight=Ducati

had a method---sound in theory---which he suggested would out perform T/T a mechanical system over 3 yrs. His method ceased to compete in 2 yrs.
So I would argue that Fundamental traders and investors are as blind as those trading technically without a proven methodology.
More than happy to see another exercise like the one duc set up 3 yrs ago.
The only genuine live methods I have seen turn consistent profit are those which have been validated.

I think people need to realise there is more than one way to invest in the market, and as a generalisation it seems funny how many mech traders think that their way is the one and only and that everyone should develop a mechanical system.

Missing the point again.
I dont care how people invest in the market or trade the market,those who dont know WHY their method COULD be profitable are just plain gambling.
Regardless of whether they have a trading plan of a few conditions or one with a checklist like a NASA launch.