How to use demo trades for better result?

[lawrence-lugar]

I'm just accustom to the term "edge".

I have seen my share of those with Skill/Experience/Wisdom that have not been able to use all that today.
They are stuck in what worked 20 years ago, not what works today.

The word 'Edge' in trading makes me think of some kind of unfair, technical advantage...like a hunter using a GPS chip to hunt an animal that's been tagged -- or insider information.
Or Edge used by marketing Guru/system vendors.

The people you mention in your example who have/had 'edge' are the prehistoric floor pit traders who were simply scalping/arbitraging their client orders for peanut profits.
I would hardly call them skillful/wisdom/experienced traders.
Most floor/pit traders are bonehead, meathead neanderthal clerks who just process orders. -- they are not paid to think.

A true trader trades his/her own account, and ideally grows it largely or expontentially. Makes dynamic, snap judgement calls in the heat of the moment. And lives to tell the tale for another day,
Cowboys! The Wild Wild West 1999...there was a giant spider in that movie. The SPY. The height of the boom.

 

[SimpleMeLike]

I think that what you do is not the best solution. Programming and watching the results can be wasting time too.
After you got the results you should analyze manually were it went good or where it went wrong. If you always test programs you will never see nor understanding the underlying powers that work. It would be like optimizing your parameters for past data and hope that it will work for future data. So you will be trading blindly without even realizing it.

Thanks schweiz,

I agree with you this on part. And that is good and clear approach and good response

 

[SimpleMeLike]

The word 'Edge' in trading makes me think of some kind of unfair, technical advantage...like a hunter using a GPS chip to hunt an animal that's been tagged -- or insider information.

This is my understanding of Edge as well.

 

[SimpleMeLike]

My personal experience is that many traders are stressed/frustrated because they have some preconceived notion of market behavior that does not reflect real market fundamentals. This is the root reason for their frustration because they build market system based on bad ideas.

A simple analogy is in the 1400s people assumed the earth was the center of the universe and build many mathematical systems that tries to explain astronomy based on the (wrong) assumption that the earth was the center of the universe. These systems while work in one area would not work in another area and as a result they became more complex over time.

Food for thoughts.

Very good explanation. I can definitely understand that.

Lately, I have taking a step back to try and understand market context/condition daily, then move into action with a trade. Afterwards, build a system based on this understanding.

 

[SimpleMeLike]

I think you need to find something that places the odds in your favor. If you don't have that, you will likely lose money after trading costs.

Bob,

What do you mean by "something" ? Some trading strategy/plan/system with high profitable stats?

 

[Wallace]

I did a NinjaTrader 7 , ES demo test on Friday morning, 7:45 - 8am pst.
I was testing 'UniRenko' bars to which I'd added a 'T3Colored' line and a 'anaSuperTrendM1'
for the arrows — nothing like overkill

A Very fast market, manual executions using REV. The UniRenko was set to 1 1 7 , a combo
amount reducing the quantities of trades with a not too large entry amount plus good trends.

The intent with the UniRenko system was to execute Buy Sell signals without reference to
Time based charts and all of the analysis I usually use.
The results were a wash so it was back to the drawing board for a new system.

The first thing one must determine and have, is a profitable trading system.
Next is Executing trades.
It's ok if the system, and/or oneself generates losing trades, as long as one Knows that at the
end of the trading session one Will have profits.
With the UniRenko and arrows system, the system was telling me what trade to make - now.
It also told me when to close the trade and open a new trade in the opposite direction, which
was executed via the REV button, producing a semi-automatic system.

The market often changes from day to day so some amount of demo trading becomes
useful for learning how one's self and the system executes within the market's variations;
how long that piece of string is will vary from system to system and person to person.

At some point one Knows the demo trading profitability of system and self; then trades live.

 

[SimpleMeLike]

I did a NinjaTrader 7 , ES demo test on Friday morning, 7:45 - 8am pst.
I was testing 'UniRenko' bars to which I'd added a 'T3Colored' line and a 'anaSuperTrendM1'
for the arrows — nothing like overkill

A Very fast market, manual executions using REV. The UniRenko was set to 1 1 7 , a combo
amount reducing the quantities of trades with a not too large entry amount plus good trends.

The intent with the UniRenko system was to execute Buy Sell signals without reference to
Time based charts and all of the analysis I usually use.
The results were a wash so it was back to the drawing board for a new system.
View attachment 182518

The first thing one must determine and have, is a profitable trading system.
Next is Executing trades.
It's ok if the system, and/or oneself generates losing trades, as long as one Knows that at the
end of the trading session one Will have profits.
With the UniRenko and arrows system, the system was telling me what trade to make - now.
It also told me when to close the trade and open a new trade in the opposite direction, which
was executed via the REV button, producing a semi-automatic system.

The market often changes from day to day so some amount of demo trading becomes
useful for learning how one's self and the system executes within the market's variations;
how long that piece of string is will vary from system to system and person to person.

At some point one Knows the demo trading profitability of system and self; then trades live.

Very nice write up Wallace.

Do you believe that a trading system first start with an trading idea the trader see and think of theirself?

In my opinion, the idea for a system first comes from manually trading, which comes from years of trading.

My only point is use paper money for the years/months of trading until a trader belive and have a profitable way to trade to make money.

Another thing I am learning as far as programming is to be selective, as one can spend countless hours programming 1000 of ideas to find nothing.

 

[Wallace]

Very nice write up Wallace.

thanks SimpleMeLike.

SimpleMeLike: "In my opinion, the idea for a system first comes from manually trading,
which comes from years of trading."
I think it would be a Great mistake to waste 'years', manually trading, or think that doing so would produce an idea.

What seems to be unique, the inspiration and originator of original ideas and programming
were the MetaQuotes Automated Trading Championships, which seemed to motivate many individuals to enter:
https://championship.mql5.com/2012/en/news/172 ~ is an interview with the 2012 winner
"I haven't traded on real accounts yet, but even on demo I was unable to control my emotions."
"Yes, I write in many languages: C, C++, Java, PHP, Perl, ASP, LISP." ! ! ! and MQL5
'What can you recommend to those who dare to engage in algorithmic trading ?' —
what about neural networks ? ? ?

There's hundreds if not thousands of trading systems in existence and it seems to me that reviewing those to see if any could be improved would be a better use of time
than years of manual trading and the hope it might generate an idea.
But what the Championship contestants seem to do was start their idea / s after deciding to enter the contest, most it appears weren't trading prior to their entry.
Perhaps it was because they were programmers, and not traders, that they started from
'scratch' to 'solve a problem' rather than design a trading system.

 

[SimpleMeLike]

thanks SimpleMeLike.

SimpleMeLike: "In my opinion, the idea for a system first comes from manually trading,
which comes from years of trading."
I think it would be a Great mistake to waste 'years', manually trading, or think that doing so would produce an idea.

What seems to be unique, the inspiration and originator of original ideas and programming
were the MetaQuotes Automated Trading Championships, which seemed to motivate many individuals to enter:
https://championship.mql5.com/2012/en/news/172 ~ is an interview with the 2012 winner
"I haven't traded on real accounts yet, but even on demo I was unable to control my emotions."
"Yes, I write in many languages: C, C++, Java, PHP, Perl, ASP, LISP." ! ! ! and MQL5
'What can you recommend to those who dare to engage in algorithmic trading ?' —
what about neural networks ? ? ?

There's hundreds if not thousands of trading systems in existence and it seems to me that reviewing those to see if any could be improved would be a better use of time
than years of manual trading and the hope it might generate an idea.
But what the Championship contestants seem to do was start their idea / s after deciding to enter the contest, most it appears weren't trading prior to their entry.
Perhaps it was because they were programmers, and not traders, that they started from
'scratch' to 'solve a problem' rather than design a trading system.

Very interesting response. I tend to agree with you as time is important. great site and manually trading consumes soo much time and energy. I can program much faster then manual trade. My problem is not many ideas, but simple ones i think of.

Are you saying that rather then trying to invent a strategy from scratch, just try to improve system that already exist?

 

[Wallace]

Are you saying that rather then trying to invent a strategy from scratch,
just try to improve system that already exist?

Why not ?
But I don't think it's that simple.
I was looking at my copy of MetaStock 11 , there's over 70 Expert Advisors - trading systems and most don't appear to produce any profit. Likewise with MetaTrader, there's hundreds of EAs but how many are consistently profitable.

I can't think out of my 'box', which is strictly analysis of the Price Time Wave Movement using my own proprietary methods, then deciding whether to enter a Buy or Sell trade based on my analysis. Unlike the Elliott Wave methods my system works because it uses irregular methods to accommodate the irregular PTWMovement.

That's The biggest problem I think with trading system design, being able to cope with irregular markets that change thru the trading session, or during a week or two of day-to-day trading.