So you want a system that actually works?

Lady,
You might read the first steps forum first.

Initially you are asking the wrong questions.

Start with how much minimum capital does it take to trade in the area of interest.
Secondly ask yourself can I afford to lose all of that without causing knock on problems elsewhere. Most every beginner looks at what they might gain first instead of looking at what they might lose and frankly the latter outcome is really a higher probability at this stage.

This should not be taken as negative. It is meant to make you change the way you are thinking about trading at this point.

Cheers
 
Jonny

Thanks for sharing your system in detail.

A slight problem I foresee when using any bar data for backtesting is the assumptions that one is forced to make about the price movement within the bar, particularly on bar one (entry bar), and also when both profit targets and stops are used.

I see 3 scenarios (for Long trades):

a) Best case - on the entry bar, the program assumes that the low price came before the trade entry, and subsequently attempts to hit the profit target before checking for a stop out
b) Worst case - vice versa
c) "Best guess" - one or other of the above, based on 1) relative positions of the bar's open and close, and 2) where the close lies in relation to the target and stop.

I think you mentioned that you were using Tradestation for backtesting (apologies if not). I haven't tried it. Is it capable of testing all 3 scenarios ? I have written a substantial amount of VB backtesting software for bar data, and the choice of scenario can make a huge difference to results.


rog1111
 
Hi Roger,

a) No I don't use tradestation I use VB and excel
b) I always but always use the worse case scenario.
c) The systems I try and design inherently don't suffer from these errors as I use 1 minute and b)

JonnyT
 
Hi Jonny

Thanks for clearing that up. I agree with you that the worst case scenario is the one to go for - I think that too many people tend to make the mistake of being way too optimistic when backtesting, as they crave a positive result. It would be interesting to know how the likes of Tradestation et al check for targets and stops, but then if they are contructing tests from tick or 1 minute data, then I guess it shouldnt be too much of a problem if stops arent too close.

rog1111
 
Can someone code this system for MetaTrader4 platform?Sadly I don't know any progamming.......
 
vladv said:
Can someone code this system for MetaTrader4 platform?Sadly I don't know any progamming.......
Johnny T has stopped trading thus I do not know if anyone else has coded it or even use it today.
 
Top