What you try to explain now is that the system uses hst data which would mean it is looking into the past. Not into the future.
It’s difficult to explain system development and backtesting to someone who hasn’t done it. Imagine iterating over data points, one by one, simulating the real world by looking at historical data. Let’s say that your simulation is starting on day -100 (a trading day that was 100 days ago). When your simulation sets the current day to -100, it should not be allowed to see the data for day -99 (a trading day that was 99 days ago.)
Anyway, when the EA Shark would trade that way forward results (good or bad) would mismatch with test results which is not the case.
I’m glad that you found the holy grail then! Can you share with us what your actual percent return has been trading this system, and over what time period you enjoyed that return? Does it approach the 180% or so per year that they advertise on their website? I’m only interested in actual cash returns from a live account, don’t bother sharing results from a demo or paper trading account.
Once again, what you write does not make any sense. Specially as i have the open version from Shark 5 and there is no such peace of code you describe.
I don’t have the source code so I cannot comment on what it is actually doing. My review was based on their marketing information and the review by the fellow from youtube. My assessment was more of an educated guess as to why their system doesn’t perform in real life as they claim on their website.
Looking into the “future” can happen in subtle ways. My understanding is that their system uses something like 10 different indicators. Any one of these may be looking into the “future”. Elliot wave and advanced get are two indicators that you may be familiar with, both of which look into the future if I remember correctly.
Besides looking into the future, a system can show phenomenal results through what is know as curve fitting. Curve fitting is especially easy to do when you have a lot of indicators or a lot of parameters. Curve fitting is performed by trying many different combinations of parameters over a subset of historical data until you get a perfect system. Then after some time, perhaps 1-3 months, you again try all the different combinations of parameters until you get a perfect fit. In this manner, one can generate outstanding theoretical results, but unfortunately these same results can never be attained with a real live trading account.
If you want to post your source code I would be happy to take a look at it. Or you could post a list of all the indicators they use and I can analyze those.