And as expected it didn't work out so well in live market. Backtesting doesn't mean much. Only forward testing on live account is proof that strategy works.
Below a link to a demo account with enabled strategy EA. Unfortunately, this is a fairly short period of time, EA was also during the tests developed and changed so they are not that clean results...
Joined Jan 2011
|
Status: dump and pump
|4,055 Posts
you still haven't factored in live trading execution, ie spreads fluctuating, slippage, and it looks like you are using control points so unless this has an avg duration of at least 1 day per trade it wont make any sense.
you still haven't factored in live trading execution, ie spreads fluctuating, slippage, and it looks like you are using control points so unless this has an avg duration of at least 1 day per trade it wont make any sense.
I achieved same results in 1 year from $600 to $130,000 http://www.forexfactory.com/showthread.php?t=544878 And as expected it didn't work out so well in live market. Backtesting doesn't mean much. Only forward testing on live account is proof that strategy works.
Ignored
Exactly Nabz ! If so easy, everyone would be rich and no need to work for a day job !
{quote} im afraid it doesn't. demo offer perfect execution 0 slippage, off quote, spread spike, nothing of that. that's why live account much different.
Ignored
I have traded demo along side of live through pepperstone and there is no difference between the 2. Its exactly the same.
{quote} I have traded demo along side of live through pepperstone and there is no difference between the 2. Its exactly the same. {quote} When testing an ea you need to fund the smallest live account possible to test. No one will buy or use an ea unless its been tested live.
Ignored
Maybe if you are trading manually or avg trade duration is over 1day. But in reality it is impossible that you get the same execution on live as a ECN model. If you understand that on a ECN you must have a buyer and a seller and a ECN will find the best bid and ask price which for most because their account is so small and irrelevant will most likely get other traders or on the broker books sometimes. But once you start trading real lots and volume you will actually get bank liquidity and subject to slippage and spread variations. The liquidity is totally different. Again, you were probably trading a small newly created account and therefore saw none of the mentioned differences. Its impossible in the long run for a demo and live to be exactly the same. I can prove it also with slippage tests and spread monitors logs but I think I got my point across.
My guess is that you fit some system to a specific time frame, meaning you curve fit a system. It is pretty easy to do when back testing and does not provided very useful information. What you need to do is take that same system with the same parameters and then test it on another time period because markets vary you will likely get different results. This is where the actual useful information in back-testing comes from because you will see in what markets you can make money and in what markets you can't. Then the game becomes how you define the markets where your system will work vs the ones where it will not.
{quote} Maybe if you are trading manually or avg trade duration is over 1day. But in reality it is impossible that you get the same execution on live as a ECN model. If you understand that on a ECN you must have a buyer and a seller and a ECN will find the best bid and ask price which for most because their account is so small and irrelevant will most likely get other traders or on the broker books sometimes. But once you start trading real lots and volume you will actually get bank liquidity and subject to slippage and spread variations. The liquidity...
Ignored
Thats true but you would have to be trading million dollar accounts before any significant slippage will occur.
{quote} I have traded demo along side of live through pepperstone and there is no difference between the 2. Its exactly the same. {quote} When testing an ea you need to fund the smallest live account possible to test. No one will buy or use an ea unless its been tested live.
Ignored
live with how much $ and for how long? Probably a sub 10k account for less than 6 months Im guessing.
