tradingsystemreviews

Going back to Tresors question about how long it takes to build a system - I have a friend of mine from university who is an incredibly clever research biologist. He has a team who are investigating a gene believed to be related to asthma. He and his team are testing several approaches to finding a way to use this to improve asthma treatments. He also knows there are at least another dozen teams trying differing approaches to the very same gene. None of them know if their approach is the one that will be successful, if indeed any of them will be successful. If they knew, they would only work on the ones that would work, but without this, they have to put their effort into every single potentially fruitful approach, if they don't they may miss the one avenue that would have been successful. It is exactly the same to come up with a trading system. You come up with a strategy, try to work out all the entries, exits, risk management, postion sizing etc for all market conditions and then back test it, forward test it and probably end up throwing it away and having to try another one. Sometimes it fails because the drawdowns are too bad an the risk of wiping you out is too high, sometimes you can't trade it because it simply does not suit you and your personality, sometimes because in real life you cannot get the fills you could in back testing, the reasons are endless. The one that works for you may be the very first one, it may be the 489th, you just don't know until you find it. It sounds like you have found a style that works for you, so my advice would be to find ways to take the grind out of it - can you automate some of it with filters, screens etc so you don't have to scan the entire market to find candidates to trade? Can you write a spreadsheet that hooks into your data provider to calculate your position sizes, stop loss points etc quickly and easily? All these things make it more efficient to trade, so you can spend less time and effort doing it.....then you can start working on your second system!!!

Hi all, Firstly newbie on here as I've only just discovered this forum...I know I took my time! Anyway first post and I'm going to start with a disagreement.... In response I'd like to paraphrase Alexander Elder in his book Entries and Exits, where he points out that although he is a discretionary trader, he believes there are more successful systematic traders that discretionary traders, although he believes that the very best performers are discretionary. Now I have to say I don't think there are too many people out there who have the discipline and talent to be able to trade their A game week in week out. We are all human; we get tired, we get emotional (remember it has been shown we suffer psycologically more from loses than we gain from corresponding gains) and sometimes we just get it wrong. If you are one of these A game people, then I doth my hat to you as I am certainly not!! As we all know trading is actually pretty dull, you find what works and keep doing it, regardless of whether you are discretionary or systematic. So it seems that if you can find your edge, mechanise it and then you can either let the computer trade it for you or at least take most of the legwork out of it and let you concentrate on the most likely candidates? If you can come up with a model that works, why not let a computer trade it for you? Your John von Neumann quote even says this - "Precisely describe to me what it is a machine can not do and I will make it do exactly that" - which is exactly what systematic traders are looking to do.