Most back testing is useless

[mitch mconal]

I don't know if it useless when looking for a "half of 1 %" of an edge over yrs. How could anyone expect more?

I have no use for it, personally. I use my set of eyeballs to "backtest" over last 30 days.

 

[foroom lluzers]

Most manual back tests are also useless , because we only seek confirmation bias in the data.
Confirmation bias comes from when you have an interpretation, and you adopt it, and then, top down, you force everything to fit that interpretation,” Kahneman. Manual back tests ignore other non biased information.It is called inattention blindness.

http://nymag.com/scienceofus/2017/01/kahneman-biases-act-like-optical-illusions.html

 

[aag100]

ive come across a trader(wont mention any names) with a fair few years experience trading who always seems to be back testing and then reeling off analysis and statistics and i feel like asking him what the F for.
if he wasnt such a sensitive soul i would ask him what he was looking for,a strategy that used to work? yet he seems to be happy and want to be patted on the back and be told well done you for tabulating all this useless garbage data nevermind your not making any money by even some sort of names in the industry

 

[Dr. Toad]

Most manual back tests are also useless

Haven't read this entire thread, and although I fully disagree with the premise, I agree with this statement.

If you don't backtest an idea, how do you know it has any merit. Might as well have a random number generator pick your entries and exits.

EDIT: Read Foroom Lluzers initial post -- don't disagree with it. In fact it is one of the reasons I system trade.

ive come across a trader(wont mention any names) with a fair few years experience trading who always seems to be back testing and then reeling off analysis and statistics and i feel like asking him what the F for.
if he wasnt such a sensitive soul i would ask him what he was looking for,a strategy that used to work? yet he seems to be happy and want to be patted on the back and be told well done you for tabulating all this useless garbage data nevermind your not making any money by even some sort of names in the industry

You wouldn't by chance be referring to me? For the record, I really would only consider myself to have about 6-9 months of experience of actually trading...certainly not years. Have taken several hiatuses where I haven't even looked at the markets for many months. I wouldn't really consider myself sensitive either. So feel free to put me in my place.

 

[foroom lluzers]

ive come across a trader(wont mention any names) with a fair few years experience trading who always seems to be back testing and then reeling off analysis and statistics and i feel like asking him what the F for.
if he wasnt such a sensitive soul i would ask him what he was looking for,a strategy that used to work? yet he seems to be happy and want to be patted on the back and be told well done you for tabulating all this useless garbage data nevermind your not making any money by even some sort of names in the industry

A friend of mine back tests for the last 15 years , every week he has a new system or modified system.He can never follow systems or existing systems due to psyche , as most traders also can not follow them.

This is also why internet has 10,000 free systems on google search.All these free systems came from bucket shop money indirectly and also amateurs on forums.

With manual forward testing you can learn new things.

 

[Dr. Toad]

With manual forward testing you can learn new things.

And also lose a lot of money if you have no properly developed and tested system in place.

 

[foroom lluzers]

And also lose a lot of money if you have no properly developed and tested system in place.

You can do it on demo , but without real money on the line , it is without real emotions and psychology.

 

[aag100]

Haven't read this entire thread, and although I fully disagree with the premise, I agree with this statement.

If you don't backtest an idea, how do you know it has any merit. Might as well have a random number generator pick your entries and exits.

EDIT: Read Foroom Lluzers initial post -- don't disagree with it. In fact it is one of the reasons I system trade.



You wouldn't by chance be referring to me? For the record, I really would only consider myself to have about 6-9 months of experience of actually trading...certainly not years. Have taken several hiatuses where I haven't even looked at the markets for many months. I wouldn't really consider myself sensitive either. So feel free to put me in my place.

No not you. someone else who has around 10 years experience or maybe even more.
And yes if you back test you will see if a strategy once had merit or not but don't bet the farm in it going forward

 

[Solas0077]

Yes but what is a changing market?

Good question. One example is the crush in volatility recently.

 

[foroom lluzers]

most manual back testing is selective attention testing , hence useless.

 

[foroom lluzers]

 

[NVP]

Back testing has merits to warm up ideas and shake out bugs ...then you gotta go forward testing to assess systems properly