Try My EA! Need Group Feedback...

See PM

Also, just using standard Steady Gold BET v2.11

Happy for you, dambuster! Good luck going forward.

A few questions, please:
- which broker are you using?
- any comments on slippage (I presume you're running it on gold)?
- are you running it between certain times or round the clock?
- any manual intervention?

Many thanks.

You're right about that, Pipster... 334% gain in under a month (with no positions open atm) is not to be sniffed at, lol!

The original one is looking poor though; all sells triggered, only 7 buys left. The only way out is down, but if gold is going to continue north, the end is near.

Good luck with your $10 test.

FXPro (was Alpari until they went bang and lost me ton of $$$ )
No slippage so far (unlike Alpari which slipped like mad)
Running on XAUUSD
Running 24/5
Manual intervention: only near the close yesterday as I didn't want trades open over the weekend and I also want to start with Steady Gold BET $10 grid v1.2 on Monday

I've been working on some other things but I didn't forget about this thread. Here's a fix for the instability in $10 grid v1.2, where it was overwriting the grid any time it re-initiated. I hope it works, I've tested it and it seems stable, but of course you need to verify this for yourself if you decide to use it. In demo mode for sure.

Some things have changed to get this version to work.




Added stuff:

PendingBuffer - this is how many pending orders that will be added to each side of grid. The EA will add enough to maintain this number of pending orders at all times. I don't think it will work if you have less than 3, it might mess with a check routine in the EA. It should be a number that will let it react to sudden moves, and would have to include some thought about the GridStep. With GridStep at 50 and PendingBuffer=5, you would be able to have to have a one tick move of $2.50 in the price to swamp the orders.

UseTradingHours - I've redone this feature to make the EA only set a grid between the hours given by the starting and stopping hours. This should allow you to keep it from starting a new grid in the off hours when significant movement is less likely. The hours are server time. Default is false - it trades all the time.

I did clean up the code to make it run a little faster. One thing I noticed in backtesting was the occasional orphan that messed up the grid, and I try to trap for that. I tried it out a few times live, changing parameters and turning the platform off and on and it didn't overprint the grid.

Like the other $10 grid EA's, you can use it to trade like the standard BET version 2.11 by opening the Sensitivity to be more than half the distance in dollars between the lines given by the PriceGrid variable.

Good luck trading! I'm running this version now on my TradeExplorer, opened up to trade like BET 2.11.

Thanks dgodfrey!

Q: can the pending buffer feature be switched off, perhaps setting dynamic grid false?

i already run the $10 grid v.13 since 2 hours ago but still no open pending post at the moment, is it normal?

@ dambuster - yes, like the other one, setting dynamic to false will force it to use the NoOfSteps setting instead. I hadn't really thought to test this, but I didn't change that part of the code so it should work. Or you could set the PendingBuffer to equal the NoOfSteps setting, which would always leave that many pending trades.

@ periphery - depending on the setting for Sensitivity-- if you have if on default settings it will have to cross a $10 even price to initiate. The last was around 6:00 or so at $1280. If it did cross that and not initiate then there's something wrong.

As always, you are a gentleman dgodfrey

Final question, am I right in thinking that in attaching the v1.3 over the v1.2 it will not take over the exisiting trades but simply create a new grid?

I'm not sure about that one. If it's got the same unique number it shouldn't overwrite the existing trades, as it sees them there. But I haven't verified that.

Thanks for the fixed version, dgodfrey...if I can scrounge up some time to update my VPS with it, I'll put up a new trade explorer with it.

http://www.kitco.com/images/live/nyg...07787682455684

Many thanks for all your hard work, dgodfrey!

Cheers, dambuster!

[quote=dgodfrey;8038541]I've been working on some other things but I didn't forget about this thread. Here's a fix for the instability in $10 grid v1.2, where it was overwriting the grid any time it re-initiated. I hope it works, I've tested it and it seems stable, but of course you need to verify this for yourself if you decide to use it. In

first of all thanks...

what time frame should be used to run the EA??

It runs on any timeframe, I've got it on 15m. It only looks at prices.

Is your v2.1 explorer the one you put the v1.3 version on? or is it the v1.2, as far as the name?

It's on $10 bet - I didn't know how to change the name. It started on Bet V1.2 but I've migrated it around with the changes, and commented here when I did. The settings on it are going to make it behave like the version 2.1 BET though, with the sensitivity window open to make it trade all the time.

One thing to remember about doing that, though, which I just found on my own as well.

If you open the sensitivity window up to make it trade a like v2.1, you will have to alter MinGridGain to be more than your target value or it will cycle out early. I just noticed my own program has a Target of 1%, which translates to around $210 currently - but MidGridGain was set at $100 and it's been cycling out on that instead. The reason is that if it was used properly (looking for the $10 lines) it would take the minimal MinGridGain to recycle on the line if it was available. But with it wide open like I have it, it is now always cycling on the smaller number.

To fix that, just set MinGridGain to a large number if you use it with the Sensitivity wide, and it will cycle on the Target number instead.

So to help someone as thick as me, if mingridgain is set to say 150, but target is say 200, it will jump out at the next $10
line if 150 is hit or close irrespective of P/L at the next $10 line?

I ask as it cycled this morning on a $10 line but ended with a small loss (am away so cant check if it was due to slippage/fast move or slow closing)

yeah, if it's >= MidGridGain when it sees the next $10 line it will get out short of the target; it will exit at above the MidGridGain. That was a valve to let you reset on the $10 line even if the target wasn't reached, as long as it met some profit expectation.

The trick is, if you have the Sensitivity wide open to allow it to reset at any time, it will always exit on the MidGridGain, because it interprets that it's always at the $10 line. Just a quirk that comes into play if you are making the EA mimic the BET v2.11 like I'm doing on my main demo account.

I think I have the default on MidGridGain set at $10, so it might have had a small loss due to slippage if it was showing a modest profit.

on my 14th cycle today on the main account : )