I have thrown together a quick version of a ZN strategy.

Below is a taste. This is 3/15 to 6/1, I am not even sure of exact dates of ZN contract periods (I need to research this).



Sharky turned me on to the ZN. I have to say, it is quite different than the ES. I am going to start paying closer attention to it.

I will share the strategy code soon. There are some things I don't want to share just yet in it.



Mike

you might try a 11 range with the 12 zerolag 12 zerolag and just buy on green and reverse on red i bet it would be very profatble dont buy and sell on the bars but on the ma itself...sam/sharky

and set it to 20 contracts if you could everyone puts in the stratiges trade on account limit or something like that i dont know what it says exactually but i would be interested to see a setting of 20 contracts
...sam/sharky

I think 3 months is not enough time. Anyone who tested ES in the past 3-4 months would have a 20.0 profit factor if they just bought the open and sold the close.

Post results of 1 year.

cunparis 

I think 3 months is not enough time. Anyone who tested ES in the past 3-4 months would have a 20.0 profit factor if they just bought the open and sold the close. Post results of 1 year.

Yes 3 months is not nearly enough time. 1 year minimum, 200 trades minimum, same as I've said before

But, I don't have 1 year of tick data for ZN because I just started looking at it, and I use Zen-Fire so it's about impossible to get it with NT 6.5.

Mike

Big Mike 

Yes 3 months is not nearly enough time. 1 year minimum, 200 trades minimum, same as I've said before But, I don't have 1 year of tick data for ZN because I just started looking at it, and I use Zen-Fire so it's about impossible to get it with NT 6.5. Mike

Would your strategy work with 1min? Do you really need tick? It looks like your avg time in a trade is 20 minutes. I think if your idea is valid it should work on several timeframes. It's rare for a strategy to only work on one timeframe, more often than not it means it's overfitted.

I could export tick data from tradestation and you could import it into NT..

cunparis 

Would your strategy work with 1min? Do you really need tick? It looks like your avg time in a trade is 20 minutes. I think if your idea is valid it should work on several timeframes. It's rare for a strategy to only work on one timeframe, more often than not it means it's overfitted. I could export tick data from tradestation and you could import it into NT..

I agree with you for the most part. I need to check it on minute data, I don't think I did already.

I've always had terrible times importing data into Ninja... have you tried it successfully?

Can you access ZB tick data back to say Jan 1 2008? That's cool if so.

Mike

cunparis 

Would your strategy work with 1min? Do you really need tick? It looks like your avg time in a trade is 20 minutes. I think if your idea is valid it should work on several timeframes. It's rare for a strategy to only work on one timeframe, more often than not it means it's overfitted. I could export tick data from tradestation and you could import it into NT..

It was a great idea but I'm not subscribed to CBOT data on Tradestation.

Big Mike 

I've always had terrible times importing data into Ninja... have you tried it successfully?

I've generated my own data from other sources (synthetic data, excel, etc.) and imported it just fine. I've never imported data exported from tradestation but I don't see why it'd be a problem.

Yeah another reason NT 6.5 sucks... contract rollovers, merging doesn't work... so I can only backtest one contract period at a time...

OK, 1 minute data here we come. First thing I notice, way too many trades. Because minute data has so many whipsaws.. I prefer volume or range... would have to add logic to script to deal with that, or just use volume as it was intended...

ZB 06-09: $5484, 600 trades
ZB 03-09: $2078, 362 trades
ZB 12-08: $6156, 948 trades

That's as far back as Ninja will let me go. So it isn't even profitable at that amount of trades, but at least it didn't completely blow up all over the place. I think with a few tweaks it could better deal with minute bars, I really need volume because it's one of the primary indicators I am using.

Mike