There is nothing wrong with backtested results. The problem arises when the data set that you are backtesting is also the one you developed the system on. So you end up with a system that can make money trading data it can "practice" on, kind of a Groundhog Day type deal. Obviously the proper way to test a system is to test it on separate data. I have developed many systems that looked great on three months data but bombed on longer periods. A lot of people here say backtesting is useless, but I disagree. If a system can't make money on historical data, I'm surely not going to risk money on it in the future. Isn't that worth knowing? It's not necessarily a guarantee that the system will make money in the future, but it's better than having no data at all.
I concur. Backtesting provides a necessary "reality test" as long as one remembers that success in the past has a substantially depreciated value for future, yet, one still can't do without its contribution. I certainly wouldn't use a system that wasn't tested. Some systems use static components and others use more volitile components that follow trends and produce too many signals. Some systems use double filters and probability tests on those frequent signals in an attempt to refine actionable signals. These systems are usually called "neuro" based shell trading systems. Some systems go to three or more stages and further filter their signals. So complexity like these certainly requires extensive and very long duration backtests, certainly greater than 3 months data, and probably greater than 3 years data.
Let see how this thing work. You don't want backtest results. You only want real time real trade result and the way to do it is that guru must really be trading the system and have real account statement to show you over a period of time. Assuming there is such a person out there who purchase the system from ABC vendor. If this person is already making money with the system why would he bother himself to tell you just for a mere $100 consider he need to go thru scanning or faxing you his account statement. Too much of a trouble for $100. And also assuming such a system really exist it will not come cheaply thou and how much are you willing to pay? $100 again? And assuming you purchase the system does it suit your personality what if the system win 1 time out of 10 success rate is 10% but that 1 win is $1000 why the 9 lose is $10 each can you take such a system? After so many years of reading and searching I come to the conclusion that holy grail does really exist and it is free if you can find it(all within yourself) and for system I have learn to develop it 99% of the system that is available is usually junk or the win/lose ratio is too low and you need to be able to trade it emotionally
what a strange thread .... is this the final egde of naive thinking or will it be followed by another thread of that kind ... what's that phrase? ... get real. peace