Brad's Emini thread

[Doji3333]

Took a little loss today..

Welp, the system took a little loss today.. Nothing to worry about though. The new stop is doing exactly what I expected it to do - limit losses.

ES = -2.50 pts = -$150
YM = -28 pts = -$140
NQ = +6pts = +$120

Oddly though, after 10AM pst, the market totally changed direction. Might have to review the historical data for this type of market reaction/action. Might be able to address this in the near future.

Still looking good though..

[Doji3333]

Another minor loss today..

yup, another minor loss today. Nothing serious though. This loss is the result of the new stop I added about 10 days ago. Might rethink this a bit- but in the mean time will let it run..

ES = -2.25 pts = -$112.50
YM = -25 pts = -$125.00
NQ = +2.25 pts = +$50.00

Net = -$187.50.

Now, if I adjusted the stop based on the trading range for the day, we would have stayed in all the trades till nearly the end of the day. Not sure I'm going to do this - but might consider it as an option.. More later..

BTW, this puts us at about +17.5 pts for the month in the ES.

[Doji3333]

Made a little bit back today..

Here are todays results..

ES = $75
YM = $75
NQ = $85

Not a KILLER DAY - but still profitable.. The funny thing about these last two days is the pre-market move. These types of pre-market activity tend to "take away" from the daily volatility.

The average (IMO) range for the ES is between 8~12 pts per day. The last two days have been just above the minimum avg daily range.. Oh well...

The good news is the system is still holding it's own.

More soon.

[Doji3333]

Net for the Week..

The NET for the week was...

-1.0 Pts for the ES..

The new stop system prevented a couple of runners but generally it held it's own while trading.

So for the month we're sitting at.. +23.75 Pts for the month.

I'll try to post some charts up this weekend. I'm pretty satisfied with this and there were no BUGS this week (at least the end of the week). So this code should be ready for others to use...( hint hint)..

B

[Doji3333]

Update..

As of today (tuesday 8/22), this week has been a non event for my system..

Yesterday (8/21) - no trade
Today = -1.5 pts = -$75

The system will do this and it is nothing to worry about. Last week, we had alot of pre-market moves that resulted in moderately flat days. This week (well, yesterday) we had a very tight range with no real direction - so NO TRADE was a good thing.

This AM, my system went long, then got stopped out. Oh well.

My assumption is the market is transitional and thus we are getting potentially volatile price swings. After last weeks pre market moves to the upside, taking a minor loss early this week is not a big deal.

The system will start profiting again when the market picks a direction and starts to trend.

[Doji3333]

New realization..

I just realized today would have been a GREAT "alt" day in my system..

"Alt" days are "alternate trading style" days. It is coded into the system and these typically happen on certain days (tuesdays not being one of them). But today there was an interesting fact that COULD be used to our advantage..

Part of my system tracks the opening range (which is a rather complex series of functions). Today's opening range was exactly the same as yesterday's. Now, this might be an isolated incident that does not play out well over the history of the symbol (ES). But it might be something that can be used in the future.

If we had traded as an ALT day, the results would have been..

+ 11 pts for today.

Now, I'm not ready to get this setup into my code yet, I have to identify IF and WHEN other similar days happened and what the outcome was. If I determine that historical data confirms my theory, I might change the code.

One thing about system development is TEST TEST TEST. Often you find little gems like this that MAY help in the future.

Still, over all the system is holding its own and we'll see how the end of Aug plays out.

More soon.

[Doji3333]

In/Out before 10AM PST (almost)

Well, the system got in and out of two symbols pretty early today (woo hoo - Golf day - I wish). More coding to finish.

ES +2.5 pts = $112.5
- Still in the ES trade
YM +21 pts = $105 (in-n-out before 10am pst)
NQ +11 pts = $220 (in-n-out before 10am pst)

I'm going to complete the backtesting functions this week. I've been doing all the historical analysis by hand and have almost finished the functions for backtesting. Yeah, I know - you program - it should be done by now. You're right, but I've been rather busy..LOL

[Doji3333]

Update..

Just got out of the ES trade with 4pts profit ($200). So not bad today...

ES +4 pts = $200
YM +21 pts = $105 (in-n-out before 10am pst)
NQ +11 pts = $220 (in-n-out before 10am pst)

Still working on improvements - but very happy with the way this is going. Hoping to finish the backtest before the end of this week.

With any luck, we might see this month gross on the ES a bit over 30 pts.

As of today - we're sitting at 26.25 (gross). So if we deduct slippage and commissions, probably near 20 pts.

[glay]

Awe man

> I'm not going to give away this (or any) efs scripts that are designed as a fully automated trading system.

[Doji3333]

Update..

Well, the reason I've been so quiet as of late is because of my work schedule (and I also found a bug in my code). This bug was in the internal calculations of the system and I thought there might have been a big problem. But after I fixed it and re-tested ofer the past 60+ days. The results are in line with my previously posted results (thank goodness). I would have hated to think a stupid mistake could cause this system to become unusable...

Well, the other thing I'm considering is to alter the stop again. The earlier version of the code included a bit more risk than I wanted (or liked). This version (with the new stop) has greatly reduced risk, but also misses some moves that would be good winners. So, I'm trying to come up with a middle levels stop system that reduces some risk while letting the trade continue.. That is next on my plate..

Oh, and I came up with another potential system over the weekend. I still have to test it and finish it. It is something I've kinda been playing with. I think I'm going to build it into a short term scalping system with the option of switching it to a trend following system. In fact, I think I can automate that process by using a simple mechanism..

Wish me luck..

[glay]

I've just written and tested about 4 EFS AT systems. They all produce losses. Its now about 5am and I think I'll get some sleep. Let me know if you want someone else to test yours out as well. I'm having a hard time finding a way to extract money from the market but I'm constantly learning on this forum. It will come sooner or later.

If you do end up posting/selling it you could make all the risk options configurable. Keep everyone informed on your progress. Good luck. Going to get some shut eye.. ttyl

[Doji3333]

Update...

Well, I just created a system that looks good. I was looking for something that would compliment my other systems and came up with a "trigger mechanism" in about 7 days. But I needed to fine tune it because it showed losses too.

You see, the trick to a trading system is..

a. focus on the faults (find out where your system generates losses and work to resolve those issues). We expect the system to make money and catch trends - right? Then the PROBLEM is the unwanted losses.

b. Develop a reasonable loss containment system. Like I tell my clients.. "It's OK to bleed a little bit with your system, you just don't want to bleed all over the place and DIE"

c. Is your system better as a Long, Intermediate or Short term system? So you've created your Trigger system and the signals look good - eh? Is it something that is only designed for one type of trading or something that is universal?? Normally, a system you develop will only be viable for certain time-frames.

d. Do you need to use Profit Targets or just a trailing stop?? OK, so Profit Targets sometimes leave profits on the table and sometimes trailing stops get you stopped out too early.. But, these mechanisms can play a very imporant role in the development of your system.

I tell my clients all the time "If developing a fully-automated trading system was easy - everyone would be doing it and everyone would be making huge profits".

The facts are these..

a. it's not easy.
b. there are lots of things to consider when developing your system.
c. planned development is much better than guesswork.
d. focus on the problems within your system (losses)
e. experience counts. If you are not successful at first, don't give up.

I've been in this industry for 17+ years. I knew very little when I first started. I learned - picked up information and developed programs for clients. Experience counts...

Anyway, back to my new system. It is not ready for full auto-trading yet, but it looks good.. 18.5 pts over the last 15 days (with some days generating no trades). I believe the total # of trades was a bit over 30. I'll run longer tests over the weekend.

So far today.. +2 pts with only one trade.

[FX989231]

Brad

I have not seen any updates lately, are you on holiday or just busy?

I always follow your work and comments - find them very valuable.

Have you tested your systems on any other markets than those mentioned in this tread?

If you remember I trade only currencies and would like to get something automated some time.

Keep up the good work

Regards

[Doji3333]

Hey there...

What can I say.. Work work work work.. If it's not one thing it's another..

My wife and I are trying to finish some improvements to our home before the first snow hits. I live 7000 up on a mountain in SoCal (www.bigbear.com) and we needed to finish this stuff cause I hate working in subfreezing weather - lol..

The good news is the Forex systems are still testing well. Of course there is still the drawdowns to consider, but this system has proven to work on most of the major currency pairs very well. So, I have not messed with that one much.

The first ES/NQ/YM system (one trade a day) is still doing well. It took almost -16 pts loss last month in the ES, but has pulled out of it without much problems. The Plumber may have been onto something with the relationship of increased volatility and the abilities of my system..

My newest system is the one that's thrown me for a loop. It seems to do well, but has it's losses too.. Tonight I may have come up with something that could greatly increase the abilities of the system. Still trying to get the coding right..

The good news is this new system seems to work well on a basket of stocks too.. (mostly on 10+ minute charts). The thing is, on ONE STOCK - your chances are minimal that you'll be profitable every day. Across 10+ symbols, then your ability to ride out intra-day losses seems to be much better.

So, to make a long post a little shorter. I'm still here. My programmers and I are working on the new version of our automated trade execution tool and once that is done, then I'll be able to BETA TEST my ideas with clients. My intention is to develop these systems for my own use and to help other clients.

It might seem foolish, but I've spent 3+ years building a suite of tools for esignal (and others) that I found necessary for my clients. One of them facilitates automated order execution. I decided I would try to build some good tools that I could offer to my clients. So, if you want to try to auto-trade (and use my tools), you don't have to spend months/years and thousands of $$$ trying to build your own to get started. You could just choose to try/use mine.

I hope my systems continue to spark interest in auto-trading with esignal - as that is what it is all about.

[glay]

I've been successfully auto trading cash with eSignal/IB for about 2 weeks now. I haven't really had too many problem except for when the CBOT went down. I'd definately be interested in testing out some of your ideas. Keep me on the list.

Geoff