You are correct that it either works or doesnt. But you would miss out on some very important education. When I was searching for my first system, I must have found 100 holy grails. Only later to run them live and have them lose money, and be left scratching my head. After careful analysis, I would discover why they didnt really work, and how they fooled me into thinking they worked. Each time this happened, I learned something. How to determine if a strategy was real before using real money. All these little "holes" I found, added up to a very important education in system analysis and hypothesis testing. It also led me to many techniques on finding new systems, and quickly determining what was bullshit. Now, anyone could give me a system, and I could analyze it within seconds and tell you if it will really work or not in a live market and be 99% right. The only reason I wouldnt be 100% right is that some edges are so extremely thin, it would come down to real life slippage for that particular system, and you could only find that out thru real trading. Then again, I would never trade such a system with an edge that thin, it leaves no margin for edge fade. Also, assuming I found a system early, I never would have developed the tools necessary to accurately detect when the system was fading, and could end up trading it for a long time losing money, before I realized the system had died. You have to know how to tell the difference between system fade and just another draw down. All these trading skills were learned on a long hard road to success because it took me years to find my first profitable system worth trading, which I still trade today.
I like to think of it as intelligently guided brute force ;-) Last night I ran 190,750 backtests, but they were highly targeted Oooooooh I made a juicy discovery too. Its was a good night.
Thank you tradedragon, that was a very good answer. And I agree with all you have said. Just one clarification: those 100 holy grails you found where strategies that worked literally "all the time" for "all the stocks" in the specified stock group (substitute stocks for commodities if you will)? Then, you found that the strategies did not work in real life for various reasons that you just learned. By having this knowledge you could reject a strategy before testing since you recognized a pattern that you know is unsuccessful. That takes a lot of time and hard work, no question about it. From your experience what would be your advice that a beginning 'learner' at this stage should focus on? What are the most common mistakes one makes? Or what are the signs of one's failure? (Just to thrown some at you for consideration: over-analyzing, rejecting too fast, focusing on complexity rather than simplicity, and of course giving up too soon). Damian: I tried different positioning strategies from a lump sum (100% portfolio) in and out, and also something like 33% in 3 trades if each subsequent trade confirmed the direction of the trend and the same for exit. Right now I focus on going long only and exiting at the right time. I do not combine shorting, or options at this point. Thank you.
Hi Rock, From what you are saying, I would suggest that your position sizing method may be a large part of your problem - that's before we even come to your method. The sizing method you are using is very crude and is only likely to be profitable if you are employing a long-term buy-and-hold strategy. Send me a private mail if you would like a chat. Thanks Damian
There is holy grail but it only exists until you lose. markets are constantly changing. For instance, there is more volume on some days than others, annual reports, economic reports, etc. influence the market. If you find a holy grail which works for 10 trades in a matter of a month. Don't be to sure that it will work tthe next time, because the markets are constantly evolving. Think, EVERY time you win someone loses. It is zero-sum game. That person which loses will try to change his strategy, thus effecting market. This is my conclusion, there is holy grail which you find but it comes and goes constantly, just like your lunch. Maybe that wasn't a good metaphor but whatever, you get the point. So, just keep evolving with the market, that is the holy grail!
Hi All, I'm always confused when people claim that the markets are constantly changing and evolving. I disagree with this completely. Maybe the way people trade the markets has evolved - in other words, traders are now able to trade the markets using more sophisticated computerised methods. But basically the markets have always done the same thing - they go up, or they go down. I fail to see how the markets are "constantly evolving" - people just seem to say that these days as an off-the-cuff remark to describe how unpredictable the markets can sometimes be. Thanks Damian U.S. Share Trader
Its really very simple, yet you people make it all confusing. Use the strategy that wins 3/5's of the time.
Let me give you a simple example. For 1998 until 2004 you could trade a spread between European Markets and US Markets. Basically you traded away the volatility (the gap expanded during the Euro morning and contracted once the US opened) and your risk was trend of the spread. It worked fine until the volatility contracted on all markets at which point the spread profits no longer outweighed the trend risks. The markets are the same - but they do evolve in the micro level. You find lots of examples where an edge disappears or changes - my personal strategy is to shelve disappearing edges for futures times and to tweak the edges of my robust strategies if things change a little (say the old stops don't work because things get pushed a bit harder). A futures trader
We got little bit sidetracked, so let me rephrase my last question. Thank you all for participation. I am learning a lot. Let's focus from now on, on a definition of a good strategy. What is the definition? Please try to asnwer the following questions: A good strategy : a) has to be combined with good money management system before it works (yes/no) b) works all the time at a specified market vs. statistically most of the time (always winner vs. always winner in the long run) c) Can be completely automated (yes/no). Once a strategy is developed it does not need any human input/modification in order to "work". It is left alone and it makes money (yes/no) d) Have to (!!!) utilize different trading strategies such as going long at times and also going short at other times, or a combination of options and commodity trading vs. simply going long only or going short only or call options only or put options only etc. e) anything else