Balls to the Wall, a 1 minute 1 trade EA

[winsteadglenn]

[SteveHopwood]

Steve: I was also having the Vivaldi problem. I noticed that the EA window was still showing London open hour, although you had changed it to Market open hour. I did a control-F search in the code for "London" and found a single instance, which I changed to Market, and compiled.
The funny message vanished. Could that be something to look at?
Regards, Glenn

I just did the same just to check, and only found references to "London" in the comments, which are left behind when I change actual code but ignore the comments.

Then, I did the same with the source in post 1. Same result.

So, no idea what caused the problem. Perhaps all you had to do was compile the source code yourself? Perhaps that is all Vivaldi had to do. I don't know. Nor do I care. Sorry, but as I occasionally remark: there is only one of me; there are lots of the rest of you.

I do know this: I can spend my time coding or I can spend it trying (and probably failing) to sort out the problems that you guys really should be sorting out for yourself, or I can spend it coding.

Which would you prefer?

[garyfritz]

I think it would depend on how big your SL is.
With this EA SL can be as small as 7 pips, this will skyrocket the position size, even though you have a fixed/controlled risk.

The position size is calculated to produce the desired risk given your SL, 2.5% in my case. I don't care if that 2.5% risk is produced by a 0.01 lotsize or a 10.00 lotsize, it's still 2.5% risk. (Assuming you don't get bad slippage, etc.)

Steve, I thought I'd throw the Tradestation backtester at it, since I *DO* trust that beastie. But I'm seeing behavior I don't understand. Can anybody clarify these points?

1. If I place the line at the "low of the last 60 minutes" as of the London open, then the line is by definition below the price, and thus you always buy. In the last year, I've gotten 250 buys and no sells. Am I misunderstanding? If I draw the line at the Open price at the open time, I still get over 70% longs, which is puzzling, considering EURUSD has trended down for the last year...

2. My understanding is that a Fractal is a bar with 2 higher/lower bars on each side. That didn't work well for me, since often (for a buy) the price would head down, driving the MACD and Stoch down, and then pop up -- and the MACD/Stoch conditions got satisfied on the bar after the fractal low, before the fractal was fully formed. Would you use that fractal low (basically the lowest low in the last several bars)? If you look for a fully-formed fractal, sometimes it forms ABOVE the entry price while the market is moving down to the fractal, so that wouldn't make a very good stop.

3. It makes a profit, but only if you don't count costs. I'm only hitting 30% wins and averaging $1.21 per mini per trade. I must be doing something wrong but it seems to be doing the right thing when I look at the trades. I sure don't see how the guy Erick quoted got 60% wins with this! Does anybody have a trade listing for the last month, that shows that 60% win rate, that they could share?

[slipshod]

Can anybody clarify these points?

1. If I place the line at the "low of the last 60 minutes" as of the London open, then the line is by definition below the price, and thus you always buy. In the last year, I've gotten 250 buys and no sells. Am I misunderstanding? If I draw the line at the Open price at the open time, I still get over 70% longs, which is puzzling, considering EURUSD has trended down for the last year...

Unless I'm misunderstanding what you're asking, you don't determine trade direction by whether the london open price is above or below the "low of the last 60 minutes", but by whether its above or below the price as it was 60 minutes ago.

If London opens higher than it was when that 60 minute period started, then you're looking for longs. If it drops below the low of that period & you're not yet in a trade, then you switch to looking for shorts.

[garyfritz]

That would make more sense, but Steve's PDF says "draw a horizontal line at the low of the previous hour." That sounds like "lowest low in the last 60 minutes" to me.... ?

If I put the line at the Close 60 minutes before the London open, I get about 60% longs, and the profits are worse.

[vivaldi]

Hallo Steve,

The estimated lot size = 0 on my screen. Also the estimated stop loss = 0 and take profit = 0. (DisableLineUse=true, that is why I see them now outside market open hours. If I set DisableLineUse=false, these estimates disappear.)

However, the estimated lot size should be something, at least 0.01. Empty4 runs in demo, with a $5000 balance. RiskPercent=5, running version 1d.
Do I have something wrong with my settings ?

Yes.

I reply in such monosyllabic mode to your posts because you are doing something wrong at your end, and I neither know nor care what it is.

Fact is, it is something that you are doing wrong. Please sort this out for yourself. Previous posts of yours in other threads, have persuaded me that you have something useful to offer in my forum, otherwise your name would not sport the 'Battle-scared vets' colour. Kindly demonstrate that you deserve this by doing some thinking for yourself.

I am not aware of me doing something wrong. I think there is something wrong in the code.
To start with the following : The variable 'stoplevel' gets integervalues like 20 in my demo account. Normally, in the real life accounts, this variable is most of the time zero and then there is no problem.
However, when it is 1,2, 3 or 20, it should be multiplied by the 'Point' value in order to make decent comparisons with values like 'Bid'or 'Ask' ..
But that is only one aspect of the Vivaldi problem.
The second aspect is more tricky. The Lotsize still remains zero from time to time. It has to do with the fact if there is a 'European low line' or not and if you have the use of this line disabled or not. But I am still looking for the exact cause.
Another story is if this has much impact on the correct working of the EA. I don't think so. Probably not, because the Estimated values are only values brought for your information. The EA does not use them.
Only the Lotsize might be of impact, but I will look into that tomorrow. However, if it is zero, it does not harm. If it has a value, it has the correct value and it is ok.

[slipshod]

That would make more sense, but Steve's PDF says "draw a horizontal line at the low of the previous hour." That sounds like "lowest low in the last 60 minutes" to me.... ?

If I put the line at the Close 60 minutes before the London open, I get about 60% longs, and the profits are worse.

Ah ok - I haven't read Steve's PDF as I've only dropped into this thread partway through; I'm going off Paulus's rules as outlined here: http://www.forexfactory.com/showthread. ... ost5640928

In particular, he says this:

Using your market open lines draw a line from EUR open to Euro close it hasn’t got to be super accurate it has to give the general bias of the market

In this example the bias is up so we then draw a horizontal line on the low of the Euro session

If this is broken before we get a signal we then look for a short.If its not broken then we continue looking for a long trade

By "Euro close" he actually means London open as is evident from the charts he posts lower down. The idea is that the Frankfurt opening hour will give a clue as to where London will go, that's the principle upon which the whole strategy depends.

[vivaldi]

I think it would depend on how big your SL is.
With this EA SL can be as small as 7 pips, this will skyrocket the position size, even though you have a fixed/controlled risk.

1. begin line in the morning ......

2. fractals .....

3. It makes a profit, but only if you don't count costs. I'm only hitting 30% wins and averaging $1.21 per mini per trade. I must be doing something wrong but it seems to be doing the right thing when I look at the trades. I sure don't see how the guy Erick quoted got 60% wins with this! Does anybody have a trade listing for the last month, that shows that 60% win rate, that they could share?

May I give my view to it ?

ad 1) Slipshod explained already.

ad 2) The software calculates the needed fractals , dependent of wether the bias is long or short.
So, when you intend to buy, the fractal (for the stoploss calculation) must be below the market and vice versa. The situation which you describe in 2) does not happen in my opinion with this EA.

ad 3) The 60 % stems from the original thread of Paulus. Paulus himself mentions a real life run of 25
trades. 15 were wins and 10 were losses. Hence the 60 %. But that is no guarantee that it will
remain 60 %. Paulus continues to mention from time to time his results in the thread. They are amazing: He manages to make almost 20 thousand pounds from a startingcapital of 2 thousand pounds. All this in a matter of a 2 to 3 months. With this method. One trade per day. I don't know how it will finish. I am reading the story and somewhere in the middle of the thread. Maybe, like in a novel, the end could be very surprising.

[garyfritz]

Ah ok - I haven't read Steve's PDF as I've only dropped into this thread partway through; I'm going off Paulus's rules as outlined here: http://www.forexfactory.com/showthread. ... ost5640928
In particular, he says this:

Using your market open lines draw a line from EUR open to Euro close it hasn’t got to be super accurate it has to give the general bias of the market
In this example the bias is up so we then draw a horizontal line on the low of the Euro session

So is he looking at the up-slope or down-slope of that yellow line, and taking that as his buy/sell bias? He says "In this example the bias is up so we then draw a horizontal line on the low of the Euro session," which I think is the red line at the bottom of the first chart. That is indeed the low of the last 60 minutes. When that red line gets broken (see 5th chart) he switches to sell bias. And from the looks of it, he doesn't switch back if it goes above -- on that 5th chart he goes short when it's above the red line.

He doesn't say what to do if the yellow line slopes down. I assume it's the inverse -- short bias unless you break the HIGH of the 60 mins before the London open.

ad 2) The software calculates the needed fractals , dependent of wether the bias is long or short. So, when you intend to buy, the fractal (for the stoploss calculation) must be below the market and vice versa. The situation which you describe in 2) does not happen in my opinion with this EA.

Yes, it wouldn't make sense. I'm just trying to figure out how you DO calculate the fractal for the different cases. E.g. in Paulus' 5th chart:

Fractal.gif

... when you go short, the fractal has not formed. Depending on how you define the fractal, you're going short either on the center bar of the fractal, or on the bar after the center. It takes 2 bars after the center to define the fractal. So you'd have to assume it's going to be a fractal and use that recent high as your stop.

You run into a similar problem when the market goes up (driving up MACD and Stoch) then reverses for your signal, and there's no higher fractal to use before it:

Fractal2.gif

So I think you have to use the recent highest High and assume that's going to become a fractal.

ad 3) The 60 % stems from the original thread of Paulus. Paulus himself mentions a real life run of 25 trades.

I don't have this strategy implemented perfectly, but I'm close. And from what I see so far, I think there is NO WAY Paulus got 60% wins by mechanically following the rules he posted. I strongly suspect Paulus has fallen victim to what many manual/discretionary traders do -- I'll bet he subconcsiously applies additional rules or filters and that improves his results. So nobody else can duplicate his results. I've seen it many times.

He posted that 60% record on May 6, and in the 25 trades before that I get *40%* wins. Which is not terrible -- it made 75% profits in those 25 days -- but it used some really enormous position sizes to do it. 83.3 LOTS for a $10k account! And with those huge sizes it averaged $300 (3%) per trade -- before costs. In reality commissions & slippage would eat most or all of the profits.

Bottom line... this doesn't seem like a real practical system.

[Lifesys]

Hi Steve,
I must be "doing things wrong" as well
OOTB we had the same problem as Sonny Jim & vivaldi.
Estimated Lotsize = 0.00
Estimated StopLoss = 12616.40 Estimated Loss = 0.00
Estimated TakeProfit = 37848.40 Estimated profit is 0.00

The Stoploss is simply Bid*factor and TakeProfit = Bid*factor*3

It is clearly not calculating the FracLow correctly,
as the fractal value is obviously Zero (0).

I completely replaced the GetFractals routine using the Yellow line value -

Code: Select all

void GetFractals()
{
      FracHigh =iHigh(NULL, PERIOD_H1, 1);
      FracLow = iLow(NULL, PERIOD_H1, 1);
}

and it works - seemingly correctly for now!!
I am going to change my copy to use the Yellow line value,
iLow(NULL, PERIOD_H1, 1)

Thanks for the fun.