Alpenkorp's Grid Scalping EA

[SteveHopwood]

Hi Steve,

The "coders" should replace the void FirstTradeOpportunity() function.
What should the "non-coders" do?

Download the update from post 1 - that is what most people do. Hence the term, "Latest update in post 1".

[bernard]

Thanks.

If you have replaced the function then that is all you need to do

Steve has just listed the lines that were added within the function

[Sojourner]

Gave it it's own Empty4 and threw everything at it. This should be interesting. =)

[Sojourner]

So far, made $4.41 and currently 54.96 DD. I even have the dow on this thing. =)

[billv]

So far, made $4.41 and currently 54.96 DD. I even have the dow on this thing. =)

how many pairs?
cheers

[wBear]

Hi Steve & Alpenkorps,

Thank you for this, greatly appreciate your work! I agree Martingale is ludicrous at a few levels deep & often, I find it's an account killer with strong trends much just like the last couple of weeks with GBPUSD as an example. I strongly believe Hedging (for those whom have the privi to do so) is the way to go as I switched out of Martingale a couple of months back & had no issues even with NFP Fridays.

I can only wish I could do proper back testing as the data I get out of my broker is rather bad & I don't have enough disk space in my VPS to download tick data files to support 2-3 year's worth of back testing. Despite so, I managed to get a $159.84 profit over the last 5 months' test period but with an undesirable 32.5% DD (USD1K account) for GBPJPY 15m. With a bit more tweak to my setfile (with a goal of bringing DD down to <20%), I believe we might be on to something

Meanwhile, Alpenkorps (or Steve) -- would you kindly share a setfile as a starting point for us please?

[Sojourner]

20 plus gold, oil, DAX30 ASA500 USA100 and FTse100

So far, made $4.41 and currently 54.96 DD. I even have the dow on this thing. =)

how many pairs?
cheers

[Alpenkorps]

Hello Guys,

I am sorry I wasn't more active at the forum these days, I was dealing with some family issues and a massive DD on my account.

I have tried many settings past few months to make this bot profitable over the long run. I was really hoping that this one would be a winner, and I still hope that it would, may be we need to do some more digging.

My backtest on higher timeframes made regular profit with low DD but due to the conservative MM (which is a must when you martingale at higher TF) the profit wasn't good enough, I am not risking it all for 10% a month. Besides my original idea was to scalp with some edge, and recover the losing trades.

Here is a backtest on 15M, TP 10, Slowkey 13/21 ema, Multiplier 3.

TesterGraph.gif

Started from January 1st 2011, account blown at September 15th. It did so good for 8 months, and the DD was so low, what happened at Sep 15th?

eu15.gif

131 pips straight shoot up in 15 mins? That was crazy! Unfortunately this move created a so big range thus DD, we weren't fortunate enough to cover it in next couple of trades.

Now that we have found the problem, what could be the solution? Avoiding news could be one, we do have a news module based on FF. The News on FF is not reliable at all, they are incorrect and they skip many important news. So this wont be a proper solution. Lets screw fundamentals and think technically.

According to the original rules, we only reverse our trades with increased lot size, when a candle closes above/below the setup candles open. By this rules, we are exposed to crazy market movements, because we are not doing anything until a candle closes, and by that time, it may shoot up 150 pips !! So we need an emergency "entry" plan, lets say we will also enter besides the original rules, if price hit the last swing high/low. Take a look the the chart below:

eu15.gif

I have added a simple zigzag indie, if we would go long immediately with the break of the last high, this trade would be done with profit.

This is one way to solve the problem, more ideas are welcome

I have also tracked down a few more breakdowns, and I think a few more options would allow us to avoid those points.

01. Max/Min candle size (high to low) applied for only setup candle.
02. Max/Min gap between MA's
03. A global stoploss

Also I am very interested to see another type of recovery, where we would enter a trade with fixed TP&SL, we will increase the lot size with the next trade setup if we get a loser.

Thats it guys, I need your help to hunt down the bugs in the logic, and find the best settings. When I did the visual backtest, I was amazed to see the trade setups and the way it was recovering. It has potential, very good potential despite the risk of being a martingale based strategy. All we need to do is fine tune it. Time for Brain Storming

[billv]

Hi Alpenkorps

Why don't you limit the lot size so that you don't have large size positions open?
To blow that size account would have required considerable exposure.
I know you want to grow the account quickly but slow and steady wins the race

cheers

[Alpenkorps]

Hi Alpenkorps

Why don't you limit the lot size so that you don't have large size positions open?
To blow that size account would have required considerable exposure.
I know you want to grow the account quickly but slow and steady wins the race

cheers

Using a higher lot multiplier has its pros and cons. Pros is, with a little move on your way, you get to close the whole deal pretty quickly. Cons .. 4/5 swings, you are in damn deep.

I have decided to use multiplier 3 after lots of backtesting and calculating. And it works pretty good. What we need now is an emergency entry and Global stop loss. Things like that happened once after 8 months ... got the point?