Hi there NewMicheal,
I'm a relative late starter here, too, but I say thanks for your contribution anyhow - much better than some that list their demands here and expect something for nothing
I think you've nailed it on the head when you say try to look to survive the bad period versus just concentrating on the win. I think most users of the system/concept would be made fast aware of that as soon as they start trading it.
Along the lines of what you're saying, using tactics to avoid the bad period and prolong your winning period, XMPH did previously post a Stats related project which as I understand it, is very good at finding statistic that are of practical use to affect what you're saying above.
I thank XMPH for that, though admittedly its over my head but I see it as a future project to explore in more detail myself. Thankfully Ive some colleagues with more smarts than I so will probably look to call in some favours to leverage the goodness I am sure resides with that project of XMPH's.
In short as I understand it, XMPH employs alot of technical indicators in his 100% mather to improve its robustness across a range of brokers, and after 6 months of hard slog myself I am frankly in awe of his achievements (aplogies XMPH if this brings you a raft of inquiries, but its the truth How much further improvement he will achieve care of this project I can only imgaine.
I think the use of support and resistance indicators to better guide the direction of the next trade is where I will look to further develope my EA, which presently just uses day and time based stats to start and finish trading each day, and a various volatality based inputs (based on Bollinger bands) to trigger the start, to hold trade baskets open longer and to continue trading later than the usual trade zone end time. Even using "just" the above mentioned, results in my EA having 52 different inputs so far........
I am well satified of the performance of my EA, now, in that it can produce a 2010 backtest on the EURUSD turning 20% month for under 50% DD but its the robustness across brokers that it lacks, and its also very time sensitive, which I'd like to decrease its sensitivity too. IF it could just turn the corner at the right time and be a bit smarter I'm sure it will only serve to improve the robustness, this as well as an adjusting vacum/kill zone.
They are a never ending project (EA's) to be sure and I plan to continue development as a long term project (in combination with a commercial release which increases the available resources required to develop - money and time). In the mean time its a very affective EA even on autopilot but I think it is more safely traded with some human intervention from time time. For example it may be looking to close basket of trades and needs to hit a certain level BUT sometimes there's resistance/support there and it just misses out. If at this stage it's x swings into a martingale I think its better to exit the trade for a small loss, or better yet be looking to reverse direction in advance of that level so that the support does get in the way - ie, it helps the EA trade in the right direction thus avoiding the martingale by simply picking it right more often.
Just which indicator to use for such purposes and how, is still the subject of my ongoing development...................if anyone here might be kind enough to offer up some suggestions I'm all ears and would appreciate the input from those have some practical experience in this department. I'm looking for indicators to reverse the direction rather than to confirm the trend because my EA trendfollows by design - its the times when it should reverse the next in line trade direction that it needs some help with.
Some ideas of mine would be swing/low high of X period Bill Williams fractals, maybe donchian channels, possibly major moving averages, round numbers and I think there's something also in the previous days high and low. This paragraph alone represent weeks if not months of testing and development.....and doesnt even touch on the more advanced indicators available.
Anyways - power to all the 100% with mather developers and believers (some might suggest!)
Cheers, Adam
I'm a relative late starter here, too, but I say thanks for your contribution anyhow - much better than some that list their demands here and expect something for nothing
I think you've nailed it on the head when you say try to look to survive the bad period versus just concentrating on the win. I think most users of the system/concept would be made fast aware of that as soon as they start trading it.
Along the lines of what you're saying, using tactics to avoid the bad period and prolong your winning period, XMPH did previously post a Stats related project which as I understand it, is very good at finding statistic that are of practical use to affect what you're saying above.
I thank XMPH for that, though admittedly its over my head but I see it as a future project to explore in more detail myself. Thankfully Ive some colleagues with more smarts than I so will probably look to call in some favours to leverage the goodness I am sure resides with that project of XMPH's.
In short as I understand it, XMPH employs alot of technical indicators in his 100% mather to improve its robustness across a range of brokers, and after 6 months of hard slog myself I am frankly in awe of his achievements (aplogies XMPH if this brings you a raft of inquiries, but its the truth How much further improvement he will achieve care of this project I can only imgaine.
I think the use of support and resistance indicators to better guide the direction of the next trade is where I will look to further develope my EA, which presently just uses day and time based stats to start and finish trading each day, and a various volatality based inputs (based on Bollinger bands) to trigger the start, to hold trade baskets open longer and to continue trading later than the usual trade zone end time. Even using "just" the above mentioned, results in my EA having 52 different inputs so far........
I am well satified of the performance of my EA, now, in that it can produce a 2010 backtest on the EURUSD turning 20% month for under 50% DD but its the robustness across brokers that it lacks, and its also very time sensitive, which I'd like to decrease its sensitivity too. IF it could just turn the corner at the right time and be a bit smarter I'm sure it will only serve to improve the robustness, this as well as an adjusting vacum/kill zone.
They are a never ending project (EA's) to be sure and I plan to continue development as a long term project (in combination with a commercial release which increases the available resources required to develop - money and time). In the mean time its a very affective EA even on autopilot but I think it is more safely traded with some human intervention from time time. For example it may be looking to close basket of trades and needs to hit a certain level BUT sometimes there's resistance/support there and it just misses out. If at this stage it's x swings into a martingale I think its better to exit the trade for a small loss, or better yet be looking to reverse direction in advance of that level so that the support does get in the way - ie, it helps the EA trade in the right direction thus avoiding the martingale by simply picking it right more often.
Just which indicator to use for such purposes and how, is still the subject of my ongoing development...................if anyone here might be kind enough to offer up some suggestions I'm all ears and would appreciate the input from those have some practical experience in this department. I'm looking for indicators to reverse the direction rather than to confirm the trend because my EA trendfollows by design - its the times when it should reverse the next in line trade direction that it needs some help with.
Some ideas of mine would be swing/low high of X period Bill Williams fractals, maybe donchian channels, possibly major moving averages, round numbers and I think there's something also in the previous days high and low. This paragraph alone represent weeks if not months of testing and development.....and doesnt even touch on the more advanced indicators available.
Anyways - power to all the 100% with mather developers and believers (some might suggest!)
Cheers, Adam