Would you Trade a System that failed a backtest ?

[omcate]

Quote from Murray Ruggiero:

I made the statement that , I would never trade a system which did not make money during a backtest. This is because a backtest is a best case look at the system results. I made these statements in another thread. It seems like some people don't agree with me. What does the group think ?

I would say "no".

IMHO:
If backtesting is similar to interpolation (curve-fitting), live trading will be analogous to extrapolation. The latter is much more difficult. In addition, new trading instruments, new regulations, new technologies, etc., are introduced to the financial market at a faster and faster pace, that is, the functional form of the curve is changing more and more rapidly. Hence, trading strategies based on historical data are becoming less and less effective. Even though you can come up with a highly profitable one, it may become obsolete within few months.

 

[barney_collier]

Backtests themselves don't "succeed" or "fail". They merely exist. It is up to you, the living breathing human trader, to INTERPRET a backtest

You may decide "I don't like these backtest results" and thus discard the system. But keep in mind that this is your CHOICE (and other traders might have made a different choice). The backtest didn't "fail" and the system didn't "fail". You passed judgement on them, that's all.

Sometimes it's easier to compare two systems to each other, than it is to compare one system against a subjective personal idea of "what is good and what isn't?" It's often easier to decide "I like B more than I like A", than it is to decide "I think B is good, period."

 

[bighog]

Backtesting is based on history, even then many will "cherry pick" the data backtested to "FIT". With enough experience one need not backtest to behind with. Why not just code what works for you in the HEAT of the battle, is not anything else just wishful thinking?

Neural networks even seem to be an attempt to code "INTUITION", another pipedream, but surely there are tons of good systems out there. I would bet the farm they are quite comparable in design.

Lets face it, even if the best system was published on this site it would not be followed, that is apparent daily from the writings of many in here, including myself..... :eek:

Like i quoted from a book once: The guy said to his traders: Do not get intelligent with the mkts, "JUST TRADE THEM".....

 

[Hombre]

Quote from Cheese:

If you want to establish predictability for each trading day in advance of the market Open (eg YM) then backtesting will not give you tradeable or reliable or sustainable results of any worth.

Cheese,
I am not, most likely, getting it right but ,are you saying that you know how to predict market moves in advance ?
Peace

 

[illiquid]

It's like asking would you look to buy a burger after you've walked into a Pizza Hut? Why did you go to Pizza Hut in the first place? People here seem to be getting sidetracked onto whether or not pizza is desirable to begin with, which is another question altogether. The salient point in this thread is: why would you go to Pizza Hut if not to buy the pizza??

 

[mama vivian]

would you buy an airplane that failed its
flight worthness test?