iSystems Journal

kevinkdog · July 12th, 2018, 06:08 AM

SMCJB

Thank you all for your nice comments. Getting back on topic though I'm hoping @rfolgate will share some of their insights and maybe highlight some systems they think are good?

The sounds of crickets is not encouraging

October 1st, 2018, 01:34 PM

I'm still here - and still planning on getting back to this - but for full disclosure purposes I did activate two systems today. These are not systems that I have specifically identified through analysis myself but systems that for one reason or another I have been watching for several months, and have now decided to activate.

October 3rd, 2018, 09:20 AM

Nice Start. Wish it was always like this.

October 3rd, 2018, 08:46 PM

thanks for sticking with this SMCJB, hope to see you succeed

October 4th, 2018, 11:20 AM

Tradesignalgo

thanks for sticking with this SMCJB, hope to see you succeed

Hopefully I'll get something more meaningful going

October 20th, 2018, 05:29 PM

SMCJB

Nice Start. Wish it was always like this.

Last two weeks have been a disaster. Both systems went long the Wed 10th drop, both getting stopped out, one of the system reentering the same day for a second loser! One system was long for the Tue 16th rally but didn't make nearly as much as it lost the week before. Then both systems were long again on the Thur 18th drop! This is frustrating because this is obviously as much bad timing as it is bad systems. It's especially frustrating as I specifically picked systems that didn't get smashed in the Feb/Mar drop so am surprised by their results. First system is on track to have it's worst month ever (but has had worse days), but the absolute numbers are not that significant and based upon recent performance it could easily make it back quickly. Second system is on track for worst month ever, Wed 10th was it's worse day ever and is $2 (literally) away from exceeding it's previous max drawdown.

At the request of a friend I created some dataset's for him this weekend. At the same time I refreshed my own dataset and will hopefully do some analysis in the next few days.

iSystems currently has

1276 Systems of which 468 or 36.7% have been added in the last year
Average system has made $86,969 INCLUDING backtest
Average system has lost $4,066 when backtest is excluded
Average system has lost $743 in last 3 months
Average license cost is $67/month
Average required capital is $5,966
Average suggested capital is $59,072 (which is 9.9x the required capital)
Average maximum drawdown is $17,227
Average current drawdown is $12,686 (which is 74% of max drawdown)

January 31st, 2019, 12:46 AM

@SMCJB I just read through this entire thread and there's really good stuff here.

I developed a few of the strategies running on isystems but will not endorse those here, nor do I trade those. But I do trade 3 of the strategies live. They are using 3 different instruments, 3 different developers and only 1 out of the 3 are profitable. The strategy which is down the most had it's biggest drawdown last month. I'm hanging in there hoping to recover. The 2nd strategy is just below B/E and is 1 trade away from going back into the black. October last year was a big DD which it has almost fully recovered. The 3rd is performing as good as I could hope. It's on it's 17th month of tracked trading and appears to be matching it's back-testing results.

I sure wish there were a couple of other good ones to add, but seeing your results, I'm thinking it's like finding a needle in a haystack. I'm sure the fees, slippage & commissions, put a real test on these strategies since they take so much off of the bottom line.

February 1st, 2019, 10:32 PM

I am not sure how much benefit the readers get without knowing the names of which iSystems are being traded.
Sure, some will be losers and some will be bad systems. It seems like trick is pick a system that is a consistant winner over time, and ride out the swings.

One important key would be to verify that the iSystems real trade prices are close to the posted trade log (green) of traders using the particular system with real money. If they are, then the real results should not be wildly different from the green track record.

I just statred trading the Swing 50Plus MiniDax FDXM on 1/23. So far, my 2 trades match the posted trade log within a tick.

February 2nd, 2019, 06:11 PM

dynoweb

@SMCJB I just read through this entire thread and there's really good stuff here.

I developed a few of the strategies running on isystems but will not endorse those here, nor do I trade those. But I do trade 3 of the strategies live. They are using 3 different instruments, 3 different developers and only 1 out of the 3 are profitable. The strategy which is down the most had it's biggest drawdown last month. I'm hanging in there hoping to recover. The 2nd strategy is just below B/E and is 1 trade away from going back into the black. October last year was a big DD which it has almost fully recovered. The 3rd is performing as good as I could hope. It's on it's 17th month of tracked trading and appears to be matching it's back-testing results.

I sure wish there were a couple of other good ones to add, but seeing your results, I'm thinking it's like finding a needle in a haystack. I'm sure the fees, slippage & commissions, put a real test on these strategies since they take so much off of the bottom line.

Thanks @dynoweb for the comments. Unfortunately my isystems journey was long and a little painful. I still have hope, and really want to get back to it. I think this dataset has to have value. I just need to find that value!

ericthewellread

I am not sure how much benefit the readers get without knowing the names of which iSystems are being traded. Sure, some will be losers and some will be bad systems.

Thanks @ericthewellread. My goal wasn't to have a journal of the systems I picked but to journal my analysis journey. That is why I've actually intentionally not talked about specific systems. I've also not wanted to be seen to be favoring or dishing specific systems.

ericthewellread

It seems like trick is pick a system that is a consistant winner over time, and ride out the swings.

Ohhhh ... if it was that easy... from previous

SMCJB

Highlighting how difficult systems picking is
(this is has been shortened - SMCJB)

I have 15498 samples of data, with 15 consecutive months of since tracked, not backtest, data. For now I am keeping back the most recent data and using 13924 samples. If a system has a 50% chance of being profitable each month, then the probability of being profitable 12 months in a row is 0.5^12 which would equate to 3.4 of my 13924 samples. In fact I have 4! Maybe surprisingly all 4 of these lost money over the next 3 months!

... cut ...

The next chart shows average profitability over the first 12 months versus average over the next 3 months for all ~14k samples.

So it does have a slight upward leaning regression, but pretty random. 189 or just 1.4% of the samples made money in 75% or more of the 12 month periods.

... cut ...

Consistent winners, don't appear to perform much better than losers going forward!

ericthewellread

One important key would be to verify that the iSystems real trade prices are close to the posted trade log (green) of traders using the particular system with real money. If they are, then the real results should not be wildly different from the green track record.

They have a slippage calculation that shows that.

ericthewellread

I just statred trading the Swing 50Plus MiniDax FDXM on 1/23. So far, my 2 trades match the posted trade log within a tick.

Reasonable looking equity curve. Live/Tracked performance is a completely different magnitude to the backtest though.

February 2nd, 2019, 10:02 PM

Sazon

Same here. Thanks for this thread and your contributions @SMCJB.

Same here i am late to the party, but i would like to say thanks for your contribution on this topic @SMCJB

iSystems Journal

Discussion in Trading Journals

iSystems Journal