if you know how to paint,you could go into the construction industry as a painter , once there you may after time know how to build the whole house and master most of the trades. It seems you found a tool that will make you money, it may support you long enough to learn how to trade most markets, i would always be experimenting ,trying to increase my knowledge ,as the markets constantly change and your tool may become outdated
Answer to thread starter : OF COURSE NOT! There are gazzilions of resons why, but here are 2: 1) "All markets" would mean "all timeframes". 2) Many markets aren't formed the same way. i.e. equities have a bullish long term bias while rate markets dont
It depends how you use auto strategies. If you want to let them run mostly unattended, yes indeed your strategy should work well in bull markets, bear markets and non-trending markets. But there are lots of ways to use strategies. For example, using a strategy to enter a trade then manually exiting to take profits. If this is the chosen method, then the question is irrelevant. Often, a strategy can let the computer catch excellent entry points, even though the exits may not work as well in a particular market. Just my two cents worth. Have been using strategies with tradestation for awhile now with good results using MANUAL exits. FYI
Of course it does not, however it certainly opens the very good possibility that your system is not really capturing any kind of market behaviour but is simply randomly in line with the instrument's data that you are testing it against. This is especially true if you get wildly different results from highly correlated instruments. If you get very different results between YM, ES and NQ and are not granular enough to be doing something that tick size and spread matter..I would do a hell of alot more testing before going live. On the other hand the correlation also opens the possibility that all 3 data sets are simply in line randomly...
Quote from M Jared: i have been working on my current system for 2 years and am considering taking out a mortgage on my house to trade with a 100,000 account. Probably one of the worst decisions you will ever make in your life. When you consider bankruptcy, you should realize that will SERIOUSLY impact your ability to get jobs for the next 7-10 years... In an employment market that is already difficult. I have backtested my system by hand the past 4 years and every year has brought over 200 % returns.I can live comfortably off of 100,000 a year as I... You are at step 1 or 2 of about a 20-step process, most of the steps bringing a dose of reality why it won't work as you planned... Your system will NOT be returning 200% a year or even 100% a year. Most come into this with the idea of making 100, 500 or higher %. Unfortunately, you will find out about things like Risk of Ruin, which basically say everyone will blow out their account at some point. The higher the returns, the much higher the chance of blowout. With your plans, if you trade like I think you plan to trade, I expect you to have your major blowout within 2 years. With almost 100% certainty. Your house will belong to the bank. I am not telling this to be cruel, I am telling this because of the massive regrets you are going to have a few years from now, with all the additional debts you undertook and how it screwed up your life and your happiness. It is believed 95% of traders lose their money. Most of the "5%" probably break even, make small amounts, short term success, etc. In fact, I saw a good article that said the few who are lucrative, usually have industry ties, they are not trading from a (mortgaged) home. Start SMALL for a year or two, buddy, don't screw up your life and that of (any) dependents or your future plans
Does it work OK on correlated markets like ES or NQ? If not, why (you don't have to tell me. Just think about it)? Perhaps looking into trading ES would avoid future problems with liquidity. If all goes well, you might just find that all your trouble to trade YM becomes trivial in the future due to liquidity. You could easily trade 300-500 contracts of ES without adding much to slippage (especially near the open and close). I know, what a problem to have...but worth thinking about it if you end of compounding your account. BTW, is this a day trading strategy? D
I think that the you already know that the answer to you question, put in those general terms, cannot be different from "no". But, probably, what you are really asking is if strategy should be "robust" and reliable. After backtesting (which is relatively useful), I think that accurate forward testing can help understanding better the behavior of your strategy and provide a good idea of what are the worst things that can happen when running it. You need to be prepared also for the worst scenarios, which sooner or later will occur. And in this sense, yes, your strategy must be "robust" to hold in these situations. ____________________ Tom My <a href="http://www.datatime.eu/public/gbot/2009_11_03_Tue_Wed_Overnight/default.htm" target="_blank">futures autotrading</a> journal
Sounds like your system works, so just carry on as you are trading it live and gradually increase the size? How many trades per day do you make on average? Nothing against traderzones, but in the time ive been here, its clear that he doesn't believe that people are able to make consistent gains from trading, and so his advice is always biased. I howver, know that to be false from people physically around me (as in, not just names on a message board). who make very nice money, day in, day out. (Im not one of them yet! lol) Dont let them discourage you. I think it would be even more encouraging if the system performed similar on all markets, but that is kind of unlikely. Different markets do behave differently, and some methods work better on certain markets.
Not only do different Markets behave differently, but more over one single Market does behave differently over time... Therefore a single System would have quite a hard time being successfull on ALL Markets at ALL time! With that said, one could think of it the other way around: a system which generates tremendous returns in one Market over one specific period of time and does awfully badly in another Market or over a different time period (out of sample back testing) should be highly suspected of not being consistently successful in the long run... IOW, look for a System which generates "reasonably good" (tbq) returns in one Market and does behave well in other Markets, i.e. produces returns within a not too wide +/- range compared to the returns for the Market for which it was built and back tested against. Not sure this is all very clear, but...