Thread: System from scratch!

Well, I thought i'd try another alteration.

I changed the inital stop to be at the extreme of the entry bar. So if long, the stop would be placed right at the bottom of the current bar.

This is how it went:



I still can't get that system number up. But its good to answer all the questions go something like: "What if I did ...?"

So.. after testing through the 1 year several times, alterting 1 variable (in this case exit). A system tester must choose the best of the variable tests and walk it forwards.

The biggest fear is curve fitting to a certain period of data.

So my selection for exit is as follows: Inital stop = approx half way between entry & the first parabolic sar dot. Of course you have to use logical positioning, for example - the low of the previous two bars can be a logical position.

for the trailing exit, we use a Parabolic SAR (0.02 / 0.02 / 0.2)

This tested the best in usdjpy 2008.

and now.. for 2009.



Max drawdown is still under 5% of starting capital.



I think we can really see the how this system performs here. I'm unsure if I will able to push it above 3.0 SQN, however currently this system is very workable.

Please note still no position sizing has been applied.

Any guesses as to what happens when I apply a position sizing model to this system?

Ok, Well here are the results using a 2% risk per trade position sizing model. This is pretty straight forwards stuff - take a bit of calculation per trade, if you want I can go into it - but let me know if you do and i'll throw up a post on it.






So... no money management this is a 20% per year return system. With money management 130%.

Hopefully this highlights something!

Next run i'll throw it through a different year (maybe 2007) with some VERY aggressive money management!

It's very interesting , the more aggressive the money management - the lower the returns. 2% might have been the correct setup.


I think, if you are interested, from my previous posts- you can build a system from scratch. I think in this section i'll be testing various ideas from the group.

I find it interesting when members post a "system" and its only an entry rule.

anyway- i'll keep you guys posted when i come across the next criteria to be tested

So, I guess out of curiosity - Has anyone picked up this system and given it a try (tested it?)? I'm about to go on holiday - however when I get back i'll be starting on something new, any ideas of what people want tested would be great!

NEVER LOSE AGAIN!! TheRumpledOne

If you have time, the TRO thread has a fixed entry and stop method. No word on targets or money management. Might be a good exercise to test this method and see if TRO is a trader or a rat (or both).

Yo Narn,

Im about to be on holidays for a month, but that's something I can look into - it would be an interesting exercise and will either give / destroy credibility to his work.

Thanks for the idea

Originally Posted by doczero

Im about to be on holidays for a month

You should be shot for that!

Originally Posted by doczero

that's something I can look into - it would be an interesting exercise and will either give / destroy credibility to his work.

He did call you a Yale student, not sure if it's a good or bad thing.

This system looks very nice ... I do have to ask though. ... are they any good suggestions for links to do with money management?