My Trading Log

[phil838]

just finished reading, but I couldn’t see any system being posted here to write the EA for?

Post 409 is MacGyver’s orginal description of it, and post 425 has my trading plan for the forward testing I’m starting.

[SanMiguel]

I’m a really bad programmer, but I’m giving it a shot right now. I’ll let you know if I get stuck.

Sure, no problem.
Obviously there are a few ways of doing it but one option could be to use a script to start the trade, which gives it a magic order number and then the EA can recognise which trades to work with. The EA will need to have a cut off point for when to stop creating new orders, any move to break even points, etc. …oh…also needs to be able to calculate risk according to account balance.

[s050399b]

Post 409 is MacGyver’s orginal description of it, and post 425 has my trading plan for the forward testing I’m starting.

I have re-read it, but didn’t really understand what’s going on…hmmmmm,

[s050399b]

The technique I’ve been experimenting with and testing is simply to place a horizontal line at the same place we’d normally put our pending order. [B]Then we open a 1min chart, or even better a 30s chart (available in Oanda) and wait for price to cross the line and close beyond it. When that happens we enter with something like a 3-4 pip (spread included so broker spread is of major importance) SL. That SL will correspond to something like 0.1-0.3 percent risk.[/B]

[B]What happens next is that we keep taking these 0.1-0.3 percent losses every time price retraces to hit our SL, until… until the trade starts running. [/B]For easy math, let’s say 3 pip SL and 0.3% risk -> 0.1% per pip.
Simple math tells us that after 10 pips we’re in a 1% profit and if we’re able to hold for 200 pips, that results in a 20% account increase.

This is one of the two versions of this I’ve been studying. The trick is to manage to catch real runners that more than outweigh all the little losers.

didn’t understand this part…

[s050399b]

wait a min!! breaking them down into steps…

1. identify a OSP_Buy, example, 1.50, SL = 1.50 - 3 pips
2. place order, OSP_Buy with a stop loss of 3 pips
3. check if order hit by stop loss
4. if order hit by stop loss, repeat step 2 until 10 tries is used up?

[s050399b]

Phil838, if what I interpreted is corrected, I could extend the functionalities of Position Sizer EA.zip here to do it?

http://forums.babypips.com/133505-post986.html

[s050399b]

The technique I’ve been experimenting with and testing is simply to place a horizontal line at the same place we’d normally put our pending order. Then we open a 1min chart, or even better a 30s chart (available in Oanda) and wait for price to cross the line and close beyond it. When that happens we enter with something like a 3-4 pip (spread included so broker spread is of major importance) SL. That SL will correspond to something like 0.1-0.3 percent risk.

What happens next is that we keep taking these 0.1-0.3 percent losses every time price retraces to hit our SL, until… until the trade starts running. For easy math, let’s say 3 pip SL and 0.3% risk -> 0.1% per pip.
Simple math tells us that after 10 pips we’re in a 1% profit and if we’re able to hold for 200 pips, that results in a 20% account increase.

[ Setup ]

• user to set “number of times to try”, “% to risk”, “buffer”, “SL in pips” in EA
• when EA is launched, there is a line for the user to move to his ideal pending buy/sell
• user re-positions the line, and activiate the EA (by deleting the label?)
• label changed to “EA Activated”

a) if current price is less than price of the line, Buy trigger, else Sell trigger
b) trigger price + buffer + spread (do we need to include spread for sell?)
c) check if the closing price of the previous 30 mins bar is more than the buy trigger price. If yes, do step e.
d) Check if the user wants want us to stop, if no, do step C
e) so closing price of 30 mins is more than our trigger + buffer + spread
f) send in order to buy at market with stop loss as specified
g) number of tries + 1
h) check if the order has been hit by stop loss, if it’s hit by stop loss and number of tries did not exceed user specify, do step C
i) change label to “EA deactivated”

http://forums.babypips.com/140108-post409.html
http://forums.babypips.com/140206-post425.html

[o990l6mh]

I see there are many posts, I’ll answer them tonight when I get home after work and gym.

[phil838]

Phil838, if what I interpreted is corrected, I could extend the functionalities of Position Sizer EA.zip here to do it?

301 Moved Permanently

I’m one step ahead of you. I’ve already modifed that exact EA into what I need.

I didn’t put too much work into it yet (everything is hard coded and the coding is sloppy), and I suggest you don’t either. We need to figure out what’s best rule-wise for this system, then work on a proper EA.

My crappy one will work for my testing purposes.

[phil838]

For the entry points I’m going to use the same criteria that I’d use to place NickB scalp lines, but I’m only going to pick the strongest lines. I know that might be a meaningless description to a lot of people reading this, but to MacGyver, Matt, SanMiguel and a lot of others that sums up about a page of rules in one sentence.

It just clicked in my brain that MacGyver’s original description of the system is using 4H swing highs/lows to trade reversals that go in the direction of the trend (James16 style). My original interpreation of what he said was using 4H swing highs/lows to trade continuations (NickB style).

I’m going to go ahead and test it both ways.

[o990l6mh]

At what point would you put the trade to break even? There are going to be times where the trade moves many pips in your direction but still comes back to hit the SL.
Also, at what point would you call it quits? Presumably, if each entry is 0.3%, you still want to run a total risk of 2% so you have 6-7 strikes?

If no-one else is a programmer, I can have a look at this for you…

A clever trick is move not just the SL, but the entire “trigger line” to cover the SL.

This can be done pretty aggressively after maybe 15 pips or so. That way many of the would be losers turn into BE instead.

When choosing the absolute best setups, that means that I would really like to give it a chance and so, these are just thoughts, I might be willing to bend the risk rule and risk as much as 3-4% perhaps combined with just 0.2% risk per trade. Expecting some BE trades, that would likely give us somewhere around 20 tries before we call it quits.

There are many posts and I haven’t read them all through, hope this hasn’t been said by someone already.

[o990l6mh]

When did you first get into Fx?

I started trading live in march I think it was and before that I had been demoing and learning since something like spring/summer 2008.

[o990l6mh]

Ok, here is my trading plan for testing out this system. Let me know if your backtesting has already shown that anything I’m considering will not work. No need to reinvent the wheel…

For the entry points I’m going to use the same criteria that I’d use to place NickB scalp lines, but I’m only going to pick the strongest lines. I know that might be a meaningless description to a lot of people reading this, but to MacGyver, Matt, SanMiguel and a lot of others that sums up about a page of rules in one sentence.

I’m only going to take counter daily trend trades if it’s a really, [I]really [/I]good setup with a good potential for profit. I’m also going to give every trade a 5 pip buffer to help avoid fakeouts.

Because I’m trying to use an EA to help with the entries I’m going to be using IBFX, which has slightly higher spreads than Oanda. Since that’s the case I can’t do a 2-3 pip SL, so I’m going for 5 pips as my stoploss. I’m going to risk .25% per trade, with a max total of 8 trades per setup, making a total risk of 2% per setup.

I’m actually going to be testing this on my live account so I don’t have to switch back and forth between live and demo. That being said I’m going to ignore my MM rules and enter each trade with .01 mini lots (since it’s real money). I’ll limit my trades to 8 per setup and record which trade, if any, went into profit so I can figure up what my profit would have been had I been using the real MM rules.

As for breakevens and takeprofits, I’m going to use good ol’ fashioned chart analysis (S+R lines, trendlines, candlesticks, etc), just like I would on a conventional trade.

So that’s my plan… Is there anything I’m missing, or anything I misunderstood about “Project MacGyver?”

Right now the only setup on EUR/USD is a long trade at 1.4844. There is a short line at 1.4650, but I’m throwing it out because it’s counter-trend and has a minor S+R area and a major trendline blocking any good profit potential.

From my experience, 8 trades may actually be too little.

I agree that 5 pips is about as low as you can go on IBFX, which in itself is a problem since every pip has a major effect on the $ per pip value.

I don’t want you to lose money over my crazy ideas Phil, so please be very careful and conservative.

When I’ve been looking at this I’ve come to realize that in order for this to be worth it, we need to use it only on the A+++ setups.

Ideally we want the trade have the potential to be able to run for many hundreds of pips, but at least 150-200 I’d say.

I guess the number one criteria should be to only take setups that offer us the chance to get onboard a major daily trend. To borrow the words from James16: we have to be very picky.

[o990l6mh]

It just clicked in my brain that MacGyver’s original description of the system is using 4H swing highs/lows to trade reversals that go in the direction of the trend (James16 style). My original interpreation of what he said was using 4H swing highs/lows to trade continuations (NickB style).

I’m going to go ahead and test it both ways.

Yes, but there may certainly be ways to improve this. I’m following your efforts with great interest.

I’ve tried many ways of deciding where to put the entry line. The only thing I’m sure of is that we want to AVOID placing it right at strong S/R levels or round numbers, since price will whip back and forth a lot here, causing many losers.

So, unusually, we’d want to place the line at a place where price tends to “blow through”. Easier said than done to find that level. James16 style is in my head, so that’s what I went with.

[o990l6mh]

What happens if you increase the risk to say 10-15 pips and reduce the attempts? Does this reduce the number of times you get stopped out before hitting the run?

This is almost a mid-point between the micro-losses and the 1-attempt traditional approach.

Yes, that will probably reduce the losers substantially, but unfortunately it will also reduce the $/pip value a lot and in an essence, that’s what it’s all about.

[o990l6mh]

I didn’t do the math but I was thinking that if you reduced the attempts then you’d still be able to maintain the $/pip value?

Didn’t think of that actually! Sure, why not!

Would be interesting to scroll back through a bunch of these setups and test how different SL sizes affect the number of losers keeping the $/pip value constant.

This is why collaboration is great, I hadn’t thought of that. I should have!

I’m going to look into that if you guys don’t beat me to it, I won’t have time until next weekend due to work.

[phil838]

Yes, that will probably reduce the losers substantially, but unfortunately it will also reduce the $/pip value a lot and in an essence, that’s what it’s all about.

I agree, the key to this is the large $/pip value.

It allows you the possibility of making very large returns with little risk. Looking back in my charts I’m seeing a lot of trades that would have made 20% or more on a 2% risk.

For every extra pip you add on to the stoploss you’re significantly dropping the payoff on a system like this. And the only real reason to try and reduce the losers is to help with the trader’s emotions anyway, because the math says you can make money at this with a very low win rate.

I’m more interested in doing what makes the most money, not what makes me feel all warm and fuzzy inside…

[o990l6mh]

I agree, the key to this is the large $/pip value.

It allows you the possibility of making very large returns with little risk. Looking back in my charts I’m seeing a lot of trades that would have made 20% or more on a 2% risk.

For every extra pip you add on to the stoploss you’re significantly dropping the payoff on a system like this. And the only real reason to try and reduce the losers is to help with the trader’s emotions anyway, because the math says you can make money at this with a very low win rate.

I’m more interested in doing what makes the most money, not what makes me feel all warm and fuzzy inside…

Absolutely, for the warm and fuzzy part we have cats and wives etc

[phil838]

I didn’t do the math but I was thinking that if you reduced the attempts then you’d still be able to maintain the $/pip value?

I think the best way to reduce the attempts isn’t to change the stoploss/max number of trades, but to work on refining the entries.

What if instead of entering after every SL after the first 30S (or M1 on MT4) candle closes above your entry line, you did it like this.

First Entry: Enter immediately after price crosses your entry.

Repeat Entries: If you get stopped out don’t reenter right away, but wait for the original entry condition to be met again.

Here’s an example from a couple days ago…

That’s just a random example I picked. Not all will take off after 2 attempts, but it seems that 5 or less is pretty common! And the losers typically don’t take up all your attempts either! Usually it’s 5 or so losers and then price just goes off in the other direction, so you hardly ever lose your entire 2%. This has the added bonus of making it safer if you wanted to risk more than 2%.

[phil838]

Absolutely, for the warm and fuzzy part we have cats and wives etc

Funny you should say that, since I have a cat sitting in my lap right now.