why this board isn't healthy if you trade for a living
- Thread starter akdrmeb
- Start date
The General
Guest
Of course they can-what another silly statement you make.
Backtesting is of no value-for several reasons-namely that it is not possible to predict the future based on past data.
If it was-then there would be some very wealthy people who would not dare to show others this "crystal ball".
You sound like a spoiled little child-who has been giving some "good advice" by the "crystal ball brigade".
When you join the "real brigade"-and fight some real battles-then and only then will you see how futile your "crystal ball" efforts really are.
The General
The General
Guest
Now I have to answer another silly post-when are you going to learn?
Mr. Hershey is indeed a "wise" man-as to his age I am not privy.
He has written some very factual words in relation to trading-as mentioned most will not see what he talks about-the reason is that they are not in the same league-they are in fact a million miles away-as if on another planet.
Continue to post foolish words and silly pictures-which are about as much use as your backtesting.
The General
Trader666,
When a trader fully understands what they are doing, there is no longer a real need for backtesting, or any need in fact. What happens in the main is that learners try to curve fit mechanical systems into history, which isn't really capturing the true essence and spirit of the market.
So, a proper experience of the market is not being gained by the learner, just partial snapshots, and not the full picture. You can see why so many give up in the end through failure of fully grasping the nettle. Learners who are taught by or with the correct procedure, test in the now, and learn by thier own natural mistakes.
This is my opinion of why so many traders have a certain distaste to backtesting, but it's just an opinion or view and in no way an opressive arguement.
Yours,
Dackster.
Edit: The General beat me to it, as his above post refers.
only thing dat work is forward testin, back testin is fo beginners like man above said
Yes, but it's when he strings those words together that he runs into trouble.
What backtesting can't do is prove that something is not random. If you have an idea that is not from the realm of random, you can take the ball and run with it with a backtester. 99% of the stuff included with the typical software is not only random but it's very lagging... and the other 1% might have a small element of non randomness but it will still lag. I'm surprised that people can even curve fit the usual stuff they backtest to get a winning strategy on historical data and I'm not at all surprised that they never seem to find anything that is really profitable.
If somebody can learn to trade then they might automate their work as an expert system by teaching the computer to do what they do. The downside to that is that the intellectual property can be stolen. You can put your strategy behind a whitelisting SPI firewall and never, ever, allow any other internet connection. Just that level of security is a royal pain when it comes to installation of software but wait, there's more.... If somebody with clout, like the NSA, wants to take over your ISP for awhile they can spoof your brokers servers and upload your strategy in the wink of an eye... or if they get control of an internet node they can do the same thing... and don't tell me they would not do that, when it comes to security you have to be as paranoiac as you possibly can... they would do that...
I'm saying the barriers to entry on autotrading are enormous really, intellectually you have to get a strategy, then you have to secure it and both of those things are rather large projects.
If somebody can learn to trade then they might automate their work as an expert system by teaching the computer to do what they do. The downside to that is that the intellectual property can be stolen. You can put your strategy behind a whitelisting SPI firewall and never, ever, allow any other internet connection. Just that level of security is a royal pain when it comes to installation of software but wait, there's more.... If somebody with clout, like the NSA, wants to take over your ISP for awhile they can spoof your brokers servers and upload your strategy in the wink of an eye... or if they get control of an internet node they can do the same thing... and don't tell me they would not do that, when it comes to security you have to be as paranoiac as you possibly can... they would do that...
I'm saying the barriers to entry on autotrading are enormous really, intellectually you have to get a strategy, then you have to secure it and both of those things are rather large projects.
Thanks for your excellent post maxpi, as it not only supports what Trader666 is talking about in his post, but speaks to the importance of backtesting to determine if you have a consistent rule-set which can be used to trade profitably in real-time trading.
T666 is using the backtesting of concepts to test their viability, determine whether the results you get are random or not and implement the use of probability in your trading to generate successful trades using the Law of Large Numbers (not for nothing jack hershey's misrepresentations about ScottD's performance using the minimal performance bond also show his complete ignorance of what good trading is all about). Once those parameters are determined, of course they must be traded in the present to see if those rules which were determined from past data still holds true.
I find it very interesting that none of the parameters used in jack hershey's SCT Trading stand-up to rigid backtesting protocols, nor were they implemented successfully in the real-time ATS developed by ScottD. This information speaks to the point that it is nothing more than a curve-fit, over optimized set of redundant rules ... none of which have any real bearing on how to consistently extract a profit from the markets ... CashCow indeed, ROTFLMAO!
Quote from MandelbrotSet:
Thanks for your excellent post maxpi, as it not only supports what Trader666 is talking about in his post, but speaks to the importance of backtesting to determine if you have a consistent rule-set which can be used to trade profitably in real-time trading.
T666 is using the backtesting of concepts to test their viability, determine whether the results you get are random or not and implement the use of probability in your trading to generate successful trades using the Law of Large Numbers (not for nothing jack hershey's misrepresentations about ScottD's performance using the minimal performance bond also show his complete ignorance of what good trading is all about). Once those parameters are determined, of course they must be traded in the present to see if those rules which were determined from past data still holds true.
I find it very interesting that none of the parameters used in jack hershey's SCT Trading stand-up to rigid backtesting protocols, nor were they implemented successfully in the real-time ATS developed by ScottD. This information speaks to the point that it is nothing more than a curve-fit, over optimized set of redundant rules ... none of which have any real bearing on how to consistently extract a profit from the markets ... CashCow indeed, ROTFLMAO!
blah blah blah... The only thing Trader666 is good at is posting and editing pictures of Jack. The only thing you are good at is sticking up for him because you two are best friends. Do you guys have sleep overs and tell Hershey jokes all night?
And no one can say shit about backtesting because as long as Eckhardt (Trader666's idol) says its good then that means its GOOD. According to Trader666 we shouldn't be having discussions on this site, we should only ask what would Eckhardt do.
LOL!