Nanningbob 60x90 EA

OK thanks will try it out. lots of testing to do.

My 4H trading thread has version 4 posted. I enlarged the section on trading 2.4.2 and hopefully better explained Mr. Z sixths indicator refreshes much faster than mine did so I am using it. However, it is a bit narrower on the lines. Dont think that will cause much trouble but we will see. Made it easier to change line colors and I liked that. Well good luck and see how USD interest rates do. I am not worried about them changing but rumors can make the market fly.

In the indicator i put early i add 0.03 to the calculated max price and subtract 0.02 to the price min and every thing is right.


------------------------------------------------------------------------

About my sliding version. It calculates for every bar the max and min price in the previous nBars(defaut 120). If your idea whit the idi is that it only refresh from week to week, i can make it calculate every monday or friday the lines and hold them until the next monday/friday, that way the lines have the same value during the week.

You could always expand the barcount to something greater then 120 previous bars; 140 may give you the large range. I personally have been experimenting with reducing the barcount down and using it on lower timeframes.

And yeah, your beating a dead horse on the EA thing, at least with me you are. 7+ years of trading experience now, working with and without EA's. I have never seen a long term successfull EA, NEVER!

However I still get excited about one that tests well, as I would love the chance to stop watching my trades personally and just trust the EA to manage them.

We agree on that. My viewpoint is there will never be an EA that can just run without monitoring because all EAs are indicator based. That mean no indicator can read the news or government reports that move currencies suddenly out of its range. I am really beginning to like the way you think. Glad you are on board here. As far as monitoring trades, using the 4H chart allows me to only check in a couple times a day to see whats happening. Big news days I am more cautious.

In Steve Hopwood's BOJ thread there is a news indicator, I think it's made by Scooby Doo. Here's the link to the thread: http://www.forexfactory.com/showthread.php?t=204646
I don't know if it can be of use to us here. Perhaps it can reduce the risk of a perfect storm. On the other hand you do need the market to move and retrace in order to make money.

Having said that, I would never rely 100% on any indicator. I agree that keeping an eye on the market and on the news will always be needed.

Happy to see all those good developments!

Which currency pairs are you using for the sixth EA.

Zura,
My backtests shows the EA crashes and burns in Dec. of 2008 on the EUR/USD with your suggested settings in post #3327. Are you using different settings now or do you get the same results as me? thanks

Haven't been around much lately and just catching up on the new system Bob. I like it alot but I will have to digest it for awhile.

I'm still playing IamThats latest ea and since 4-5-10 till today I'm 6200 pips up playing 16 pairs and just letting the ea run 24/5 on a 4hr tf....Dan

Are you talking about the one from post #3168? Thats pretty impressive results.

Yeah, apparently I forgot to download the history file when I re-installed MT4 the other day to get the EA working, so all my backtests were standing on Jan 9, 2009.. go figure.

So I double checked the EUR/USD during the month of december for 2008. The problem your seeing there is the price range become simply crazy from Nov 21 to Dec 15. The High to Low is 1564 Pips which is just too large a move for the system to handle with an account size of $2000. When your TP hits 225+ Pips and the distance between bars is 250+ Pips it becomes easy to margin call the account using the martingaling strategy the EA currently uses.

The problem with the EA in its current form is that "The Trend is NOT Your Friend!". This EA makes its money off of a ranging markey, which is what the market is 90% of the time IMO, and why trend trading EA's fail over time. If/When we get Nanningbobs 2:4:2 recovery plan coded in, it should help prevent blowing accounts in situations like that. The only other option is to have a larger account, for instance using the exact setting I posted and simply increasing the account size to $6000 to weather that $4000+ drawdown in December.

This is the one with the ma. I don't use that feature. All pending orders open at .01.

Although I've been a pretty avid student of forex over the last 5 or so months, I still often put my foot in my mouth. So take that into account as I throw my two cents in here about backtesting. Basically, I believe backtesting this EA--even if it had the 2.4.2 system integrated into the code--seems like a waste of time and could result in far less profitable settings.

Here's why I think this:

First, according to Bob (and I have no reason not to believe him), the mt4 backtester is flawed. That alone makes me wary of using backtesting as a means to determine settings. But also, I feel like this system needs a human-controlled trade and money management strategy present at all times to wring the profitability potential out of it.

I'm drawing these conclusions from the experiment I did over the last month using IamTHat's EA on the 5M timeframe. I ultimately abandoned this strategy as I was spending 6-8 hours a day babysitting the platform to make 10% on the account--not bad, but that 10% was a result of a relatively small margin between a lot of small wins and a handful of big losses. The bigger time frames, I feel, can be just as if not more profitable and a whole lot less stressful--although the idea of using a "driven" nanningbob EA (buys or sells only) on the 5M as price approaches the A lines is appealing... hmm...

I digress..

Anyway, I found that to even get the profitability that I did, I developed a "fluid" trade and money management strategy that changed on the fly as conditions changed. Take these tips for what they're worth considering my introduction.

1. Stay on the sidelines during news and use the 2.4.2 system: These are Bob's key rules and they should be taken as gospel. If you're using a backtest to determine settings and letting it play during news stories and not using the system, you're totally going to skew how deep trades usually go on average and way OVERestimate your drawdown and margin.

2. "Pyramid" your margin: Have a couple of trades going to the second or third level or even recovery? It's going to happen, especially in a trending market. In my experience, as trades went to deeper levels, I would slowly shut down the EA on the currencies with no active trades until I was out of the jam. That way I could be sure I could save my margin for the pairs with open trades.

3. Less currencies, bigger lot size: Think about this... what advantage does running the EA on AUD/USD and AUD/CAD at the same time have? If an event happens to effect the AUD that causes a bad trade open, most likely you'll be dealing with two bad trades instead of one. In my opinion, you might as well just have one or the other running with double the lot size. It gives you options, too: Ben Bernanke talking about interest rates tomorrow? Sit A/U and play A/C instead. In general, I think after a couple months of FORWARD testing, I'll try to cut the amount of currencies i'm trading in half to the most profitable and diverse in terms of region and risk/reserve status and just double the lot size.

4. Make sure lot size grows with account size: I'm not sure if those running the backtests are leaving a static value for lot size or letting the ea calculate by account size. Not taking advantage of the effects of compound interest will seriously limit your profitability and ability to grow the account. I don't plan to be making the same profit in real dollars on my account in December as I did in January.

These guidelines have a simple goal in mind: Keep your money as busy as possible while managing margin and drawdown. I saw someone's backtest that featured 5 levels and a .01 lot size for a $2000 account. Any greater lot size and the account would blow due to the fact that level 5 is 16 times greater than level 1. No offense, but this seems absolutely ludicrous to me; like making a one size fits all uniform based on the proportions of the tallest kid on the basketball team.

I think it's reasonable to expect that the vast majority of trades made by this ea will take profit by the first or second level. It's therefore imperative that these levels be as big as we can possibly make them while keeping risk at a manageable level. In my opinion, using rule #1 alone could allow you to double the lot size in the above example with the same or possibly even reduced drawdown and margin use. Double the lot size means double the profit. If this means having to close out a losing level now and then in recovery, that will be a small price to pay in comparison!

I feel like I'm rambling on a bit but I just wanted to share my thoughts and experiences from trading live with a nanningbob system for an entire month. To paraphrase Bob, you have to take into account the limitations of any EA or indicator and account for it yourself if you want to be successful.

*exhales*

Take care.

Hello Zurathustra,
I've setup your EA on a demo account and have been running since start of trading on Sunday evening. I've got two issues to report. For some reason, a trade was not entered on a couple of pairs when the price is clearly outside the A lines, specifically, GBP/USD and USD/CAD.

Secondly, seems like the EA will not reenter after a close if the price moves outside the A lines again, example: EUR/GBP. Not sure what is happening there. Other than that, it seems to be running fine.

Are you going to program trading on the B lines?

Nice job altogether. Thanks.

Field

Great job Jay, couldnt have said it better. Gonna cut and paste this and probably put it in one of my books. Definitely going front page, if you dont mind. Good luck to you.

Some of you guys maybe good fundamental traders but my question is this: How stable is the EURO? I am afraid it may cause another OCT. 2008. The stories of it breaking up or some countries leaving or getting booted will make the market go crazy. Anyone got some good solid facts on this issue or know where to go to find solid facts??

I usually keep my eyes on the news all the time but this is tricky. I don't think any country would voluntarily leave the euro, it offers too many benefits plus all the participants have invested too much already.

The is a possibility of a "new Greece" though. Ireland, Italy, Spain and Portugal all have serious balance issues. The thing that seems different from Greece is that Greece actually manipulated it's figures in the past to make everything seem ok. These other countries seem to have all their cards on the table. I assume their current situation is already factored into the current market. If and when it will turn out there could be another rabbit in the hat somewhere it could mean another serious blow to the euro. Obviously this counts for any currency.

In any case, news like this is hard to predict. Everyone is watching this closely so even the slightest hint can hit like a bombshell. There's no easing in to it.

Not sure whats causing your problem and the post you qouted has the info for trading either A or B so it has already been programmed. I've been running the EA since monday afternoon on all standard pairs that IBFX offers and am so far up $39.55 for less then 10 hours of trading.

Your are running the Sixth Screen EA v3.2z I wrote? It should be drawing lines, telling you the High and Low its currently working with, the Range in pips from High to Low, Take Profit, Line Pip Distance, and Sell Level Buy Level, which are the prices it will be entering trades.

Make sure your Sell Level and But Level match the lines drawn on the screen. If not you may have set your ABFlag incorrectly.

ABFlag Settings are:
0 = Trades the High and Low (Very Conservative)
1 = Trades the A Lines (Default Setting)
2 = Trades the B Lines (Aggressive)
3 = Trades the Center Line (NOT A GOOD IDEA)
4+ = Strange Behavior, I dont recommend

Thanks. I was running the prior version looks like.

I'm up 409 pips from Sunday night trading only the A levels on 21 pairs at IBFX.

Great job.

Has anyone noticed that the 3.6 indicator won't delete off of the window, no matter what you do? Weird.