I guess Nikolai s view is unfortunately clouded by his own experience and as a result not necessarily objective - not sure why he states that your strategy has no logic to work as we do not have much info on your approach. However he is right saying that curve-fitting risk is always high - especially if you have a very large number of indicator/conditions. An interesting test for you to do is the following: you should be able to explain the simple/basic market context in which your strategy works and does not work - if this is to complicate to articulate, you might have a problem. I encourage you to continue your research, serendipity is the researcher best friend ;) + trade it live, no better test. best

@Antoine,

First, I forgive you for misspelling my name.. However, I do take the time to spell your name correctly.. Hoping next time you do too.

Second, I am just cautious about strategies that work just based on data. I do want to understand where the edge comes from. Then, I can have more conviction in putting money behind it. As I said somewhere else in the forum, it is one thing to NOT know if strategy works, totally different to believe in a strategy that actually does not work. If that causes me to miss an opportunity, I am ready to pay the price and miss that opportunity.

Finally, what is really clouding my judgement, I just do not think those returns are possible in SPX without leverage. And certainly want nothing to do with breakout strategies. I assume others may believe otherwise. It is possible I am wrong but there are different strategies for different people. Good luck to all.

I believe that market neutralizes any strategy and no strategy tested on historic data

may continue to give the same profit, as markets keep evolving. Even said that, markets are run by emotional humans, whose greed and fear patterns/behaviors never follow random walk and always lead to irrationality.

@Nikolay: Some logic should be there in my conditions, unless it doesnt work for so many years. And by the way all those returns that I posted are with out leverage, if I consider 5% margin for trading EMINI, then the returns turn out to be even more phenomenal. I will take your point and will try to trade live and see how it goes.

@Antoine: Its not that I cant explain some contexts, but will try to see how the out of sample will keep working for few more months and then will post my analysis.

Some, when they see someone showing performance results above their own are quick to advance all the possible negatives they can find. They operate on the premise: if they can't do it, nobody else can.

I prefer to see anyone's trading strategy with a cup half full: for instance, I'd rather think that if I can do something, then anybody can do better. After all, how hard could it be?

There is a lot of information in @Pavan's displayed test results and comments. Some of it could cast doubts on the overall performance levels, but I think this might be due in part to the lack of additional details. For sure, a 41.3% CAGR over a 32 year period playing the S&P 500 index is most impressive. The S&P 500 over the same 32 year period had a CAGR of only 8.898%. This implies some 32 alpha points, and these points are quite hard to come by.

When compared to Mr. Buffett's performance level since 1982 (20.27% CAGR); it would classify @Pavan at the very top of the totem pole.

For now, the details provided by @Pavan are insufficient to ascertain his trading methodology much further. However, I'm starting to have doubts that could be easily resolved simply by providing a snapshot of the summary performance report generated by the simulation software used. May I suggest @Pavan provide one. It won't reveal his trading script, but it would kind of validate in some way some of the things he says.

Here would be one of my questions. From the return numbers provided and starting with a capital of $100k, the portfolio would have grown to $1.9B by 2010. And for 2011 would have made an additional $1.5B when the index went up by 19.76%. This has to imply very very large bets and since @Pavan mentioned that he did not use any leverage, it would tend to make his performance results that more impressive. Even more impressive if less that the total portfolio was put into play.

as long as you dont use LMT or STP orders (Indices OHLC bars do not reflect the overnight gaps - the bars are artefact) the results should be ok and replicable.

Entering and exiting in Close (or next bar's Open) is basically ok (better the Close, as the Open is sometimes difficult to trade).

Simply take the relevant future (backadjusted) and run your system....