Backtesting is useless

[SimpleMeLike]

That's my point you can't know that it will make money doing backtesting, you are testing on only one version of the history.

I agree that backtesting can be free practice but it is biased because you already know what the market will do. I think paper trading will be a better way to practice, or, even better, real trading with very small positions.

I disagree that improves your odds of winning, you can't control your odds in real life, you don't even know how many faces the dice of life have.

Hello JCDST1979,

Great conversation.

You are right, you can not "Know" for "100% Certain", but what if you know for 60%-80% certain? What if the back test of 1000 sample points say if you trade this way or with this idea, you have a 60% chance of making money? Would you except these odds going forward?

Look at all the the advantages in life you get with a 60-80% chances of winning with a 1000 sample data:

1. You got some good practice. You are training your eyes and brain to react without a DOUBT, without fear, without overthought, you are building your confidence.
2. You got some good confidence
3. You got some good stability.
4. You got to see something that "may" be repeatable in the future.
5. You have a test, that gives you some advantages in life of MAKING MONEY. It is one layer of support, you can trust.
6. You will be happy and not stress. Remember, you have to be happy doing this job too.

I disagree that improves your odds of winning, you can't control your odds in real life, you don't even know how many faces the dice of life have.

I disagree with you. It's ALLL about odds in life and trading. If I am raised by a rich family, my odds that I will be rich is certainly greater than 50%. If I am raised by a poor family, my odds that I will be poor is certainly less than 50%. Which family would you rather be raised by? You can reverse this logic by a person race if you like if you live in America.

You put yourself in a position to win, regardless of how you "feel" something may play out in the future. Surround yourself with smart people, or quality data. ALL I have is my

In order to win, you MUST have the data and the skills and the preparation. Back testing for FREE with $0.00 cost gives you all of this.

 

[lindq]

How can you be sure that backtesting can save a lot of time and capital?

20 years, that's how.

 

[JCDST1979]

1. You got some good practice. You are training your eyes and brain to react without a DOUBT, without fear, without overthought, you are building your confidence.
2. You got some good confidence
3. You got some good stability.
4. You got to see something that "may" be repeatable in the future.
5. You have a test, that gives you some advantages in life of MAKING MONEY. It is one layer of support, you can trust.
6. You will be happy and not stress. Remember, you have to be happy doing this job too.

I agree with you that you can get more confident in your mechanics, but I still think that for this you are better or paper trading or trading with a very small risk.

You are right, you can not "Know" for "100% Certain", but what if you know for 60%-80% certain? What if the back test of 1000 sample points say if you trade this way or with this idea, you have a 60% chance of making money? Would you except these odds going forward?

That's like saying that the past is 60-80% equal to the future, Do you think so? it would very easy to make money that way.

May I didn't make myself clear, if you backtest your trading strategy let's say in AAPL in 2019, you are testing your strategy in just one of the possible outcomes of AAPL in that year. You can't have a 1000 sample data there. You can have 20 years of data but every year conditions are different that isn't a controlled experiment.

Let's say you have a bag with 50 red balls and 50 black balls, it's a 50-50 shot right?. I agree, but what happens when someone without you even knowing it Is adding balls to bag or taking them out. Do you keep trusting that odds? How many balls are there in the stock market? Is someone adding more balls? How many? Witch color?.

I disagree with you. It's ALLL about odds in life and trading. If I am raised by a rich family, my odds that I will be rich is certainly greater than 50%. If I am raised by a poor family, my odds that I will be poor is certainly less than 50%. Which family would you rather be raised by? You can reverse this logic by a person race if you like if you live in America.

I agree it is all about the odds but you can't relive your life, you only have one shot.

 

[JCDST1979]

20 years, that's how.

https://www.thoughtco.com/appeal-to...omething that works,argument, it is a fallacy.

 

[SimpleMeLike]

I agree with you that you can get more confident in your mechanics, but I still think that for this you are better or paper trading or trading with a very small risk.

I agree. forward testing in real time give you more experience or trading with small risk. I agree. I do more forward testing then back testing myself.


I intraday day trade, So if I back test, I can get about 800 testing data samples per year. I am seeing more bars and patterns through out the day versus swing trading, maybe 50 trades a year or something.

Are you referring to day trading or swing trading back testing? For day trading, you have to back test alot.

 

[SimpleMeLike]

What about this article from Al Brooks.

https://www.brookstradingcourse.com/ask-al/backtesting/

Hello Heydrrich,

Very interesting coming from Al Brooks. I 1000% disagree with Al Brooks. I am bit shocked he is saying this, honestly, because he teaches price action concepts. So, a trader is suppose to buy his course and just start trading in real time with real cash? yeah right.

 

[userque]

In the book Fooled by Randomness, Nassim talks that the history that unfolds is just one of the multiple possible ones, this makes backtesting completely useless, don't you think?

No.

Suppose that a percentage-change price sequence is actually a collection of shorter sequences that repeat, randomly. So while the whole history may be unique, it's components are not.

So a strategy designed to capitalize on these smaller, repeating patterns, does not need the complete past history to repeat as it did during a backtest; it only needs the shorter sequences within that backtest to repeat--even if randomly.

This satisfies Nassim's assertion, while also allowing a backtest to be useful.

 

[SimpleMeLike]

He's tacitly admitting to trading without any edge, my guess anyway.

Some confusion can stem regarding backtesting of a strict mathematical/"mechanical" rule versus manual backtesting by a discretionary or semi-discretionary trader on events as they happen in a recorded stream. I didn't realize the latter existed for many years. I am not sure what Al is talking about here, suffice to say the smart people he's mentioning are doing the former. If you're doing the latter and it's not comparable to your discretionary live trading, it's probably because your environment wildly differs (for instance, knowledge about the current market regime which you won't know about if randomly dropping into some 10 min bars somewhere, or you're not replaying events in real time, or with different software than your live trading, etc.).

If I had to guess, he probably has been trying some very basic bar patterns or a few indicator combination rules, rules that machines have tested trillions of times (or more) and that generally has no edge (from what I've seen, YMMV). In his belief system (as is rather common) this stuff should have an edge, so he's disappointed.

Tangentially, I own a couple of his books that I bought very early on in a batch. He has incredibly low signal to noise ratio in his writing.

Snuskpelle,

I agree with you.

 

[_terminus_]

https://www.thoughtco.com/appeal-to-age-fallacy-250345#:~:text=Explanation of the Appeal to Age Fallacy&text=The Latin for Appeal to,the most common form is:&text=Sticking with something that works,argument, it is a fallacy.

You're running the risk of upsetting Maestro Taleb. He's fond of anything Lindy.

 

[SimpleMeLike]

I agree with you that you can get more confident in your mechanics, but I still think that for this you are better or paper trading or trading with a very small risk.



That's like saying that the past is 60-80% equal to the future, Do you think so? it would very easy to make money that way.

May I didn't make myself clear, if you backtest your trading strategy let's say in AAPL in 2019, you are testing your strategy in just one of the possible outcomes of AAPL in that year. You can't have a 1000 sample data there. You can have 20 years of data but every year conditions are different that isn't a controlled experiment.

Let's say you have a bag with 50 red balls and 50 black balls, it's a 50-50 shot right?. I agree, but what happens when someone without you even knowing it Is adding balls to bag or taking them out. Do you keep trusting that odds? How many balls are there in the stock market? Is someone adding more balls? How many? Witch color?.



I agree it is all about the odds but you can't relive your life, you only have one shot.

JCDST1979,

Ok man, stop bullshitting around man. You know what I am saying. Stop playing around man

You need to test what you are doing before committing money to it. You can forward test in sim or use market replay in back test.