Determining a stop loss

[OGRooney]

I see no reason why you would interpret my sentence with an upper class English accent, as opposed to say a Southern Kentucky yobbo accent. Nothing in my post was trying to attack you and it is unforunate that you have interpreted it as such. The same applies to a number of other posts in this thread - most are just debating the topic at hand and I can't see anthing directed at you personally.

16 trades being an inadequate sample size to determine win% longer term is a simple statement of fact. If you have a larger sample size from your months of trading feel free to put those on the table.

Yeah and show you all the losses I encountered as a beginner - that would really be a good PR exercise for my credibility on this forum. I had some dirty losses when I didn't know what I was doing as most people probably did. I shared those stats cause they were from the last few days and on hand (the account I was actively trading). You said you couldn't understand why saying "and do go get" to someone was rude or snobby so I asked you to say it to yourself in a an English accent, I used upper class to differentiate between the standard British accent and a cockney one. If you'd used a yokels accent of any country, the effect would not have been the same - I take it from your concern that you didn't actually bother going to the mirror and giving it a shot?

Eitherway, I did pretty bloody well last week. I loved every trade I made and it paid off.

I concede this round with a painful case of scoliosis, I'll see you all tomorrow.... or maybe in the new york session

 

[burglar]

... I concede this round with a painful case of scoliosis, ...

Ok!! You had me scurry to the dictionary !

 

[skc]

You said you couldn't understand why saying "and do go get" to someone was rude or snobby so I asked you to say it to yourself in a an English accent, I used upper class to differentiate between the standard British accent and a cockney one. If you'd used a yokels accent of any country, the effect would not have been the same - I take it from your concern that you didn't actually bother going to the mirror and giving it a shot?

Person A... "And do go get high tea at the Ritzs while you were over in London".

Person B... "Why are you putting me down?"

 

[prawn_86]

Back on topic please all

 

[TulipFX]

I agree, All I've been saying is this system could be made more efficient by reducing the amount of times you were incorrectly stopped out (ie times that you would have made $500 but instead chose to incur a loss)

What about when you were correctly incorrectly stopped out and only lost $500 instead of a massive chunk?

In forex we have found low risk multiple position traders with set SL/risk + exits (both profits and losses) on pullbacks are good ways to go. Rarely is our hard SL hit, 80+% of the time the trade exits when the initial conditions that had it enter no longer exist and have flipped.

If on a long, the price goes down, against you, then pops up - exit on that over bought indication. Allows you to cut many losses shorter than allowing the trade to linger and travel down to a full stop out.

Of course there are times when the trade will come back, but statistically on our systems over thousands and thousands of tests this idea proves time and time again to be the most profitable one (or least loss incurring).

 

[pixel]

I find it mildly amusing, how a debate can rage on about two sides of the same coin without either side realising their own limitations.
The complete argument of the two sides can be described quite simply:

1. Setting a stop-loss too close to actual support risks getting stopped out unnecessarily. But not setting a stop-loss at all risks going broke. The skill lies in determining the stop-loss level that is appropriate for the particular instrument, trade, and planning horizon.

2. Having been stopped out of a position mere ticks before a turnaround is no reason to stay out of the trade. There is no law against starting a new position once the tide has turned.

All too often do I find that Ego and the view "I'm wrong??? Never!!!" get in the way of making the adjustments that have to be made when market conditions change.

The Market can remain "wrong" far longer than any individual trader can remain solvent.

 

[burglar]

I find it mildly amusing, how a debate can rage on about two sides of the same coin without either side realising their own limitations. ...

+1 pixel

 

[Boggo]

I concede this round with a painful case of scoliosis, ...

You may be getting be getting the curvature of your spine mixed up with a lack of knowledge young OG.

sciolist - an amateur who engages in an activity without serious intentions and who pretends to have knowledge

I am leaning towards believing that the latter may be more prevalent based on the absence of evidence to support some strong opinions

1. Setting a stop-loss too close to actual support risks getting stopped out unnecessarily. But not setting a stop-loss at all risks going broke. The skill lies in determining the stop-loss level that is appropriate for the particular instrument, trade, and planning horizon.

2. Having been stopped out of a position mere ticks before a turnaround is no reason to stay out of the trade. There is no law against starting a new position once the tide has turned.

All too often do I find that Ego and the view "I'm wrong??? Never!!!" get in the way of making the adjustments that have to be made when market conditions change.

Excellent, that sums it up nicely pixel.

When I started out I was risk taking, then controlled risk taking, ie risking 2% to 3% of my capital on any trade. Have been through the whole ride em down, average em down etc thrills, improves your cardio capacity but does nothing for your account.

Through sticking to a plan of being prepared to accept small losses I can now more than double the size of my orders and I am only risking 0.33% to 0.37% of my total capital on any single trade.

Losing and breakeven trades are just the price of doing business (all two thirds of them !)

 

[OGRooney]

Excellent, that sums it up nicely pixel.

You agree with the guy in the thread who actually got what I was on about and summed it up for everyone? Righto then.

 

[CanOz]

You agree with the guy in the thread who actually got what I was on about and summed it up for everyone? Righto then.

Yup, that's whatcha call a "forum moment", and i seem to get most of them!LOL!

CanOz

 

[Wysiwyg]

Bulkowski has the usual stop techniques and one that is a bit different from the ATR volatility stop is this one I put together for Amibroker. The range is usually different to standard ATR(22).

In Kaufman's technique, compute the average daily high-low price range for the prior month, multiply by 2, and then subtract the result from the current low price.
Also, the look back should be 22 price bars. That is about a month's worth of price data that you average.
The average high-low volatility measure (HL) performs better than the average true range (ATR, which includes gaps, whereas the HL method does not) and better than standard deviation.

http://thepatternsite.com/stops.html

//// ATR Modified \\\\

HL1 = Ref(H - L, -1);
HL2 = Ref(H - L, -2);
HL3 = Ref(H - L, -3);
HL4 = Ref(H - L, -4);
HL5 = Ref(H - L, -5);
HL6 = Ref(H - L, -6);
HL7 = Ref(H - L, -7);
HL8 = Ref(H - L, -8);
HL9 = Ref(H - L, -9);
HL10 = Ref(H - L, -10);
HL11 = Ref(H - L, -11);
HL12 = Ref(H - L, -12);
HL13 = Ref(H - L, -13);
HL14 = Ref(H - L, -14);
HL15 = Ref(H - L, -15);
HL16 = Ref(H - L, -16);
HL17 = Ref(H - L, -17);
HL18 = Ref(H - L, -18);
HL19 = Ref(H - L, -19);
HL20 = Ref(H - L, -20);
HL21 = Ref(H - L, -21);
HL22 = Ref(H - L, -22);

RangeSum = HL1+HL2+HL3+HL4+HL5+HL6+HL7+HL8+HL9+HL10+HL11+HL12+HL13+HL14+HL15+HL16+HL17+HL18+HL19+HL20+HL21+HL22;
AverageRange22 = (RangeSum / 22) * 2;
AddColumn(AverageRange22, "ATR Mod", 1.2, colorBlue);

 

[CanOz]

Can that be done in an array?

 

[Wysiwyg]

I am sure a shorthand version is doable.

 

[captain black]

I am sure a shorthand version is doable.

This version uses "sum" to reduce the number of "ref" calls. Sum begins with today's bar and the formula begins with yesterday's bar so it's necessary to use (Ref, -1) on the RangeSum to sum the values beginning with the previous bar.

RangeSum = Ref (Sum ( H-L , 22 ) , -1);
AverageRange22 = ( RangeSum/22 ) * 2;
AddColumn(AverageRange22,"ATR Mod",1.2,colorBlue);

 

[Habakkuk]

I am sure a shorthand version is doable.

AddColumn(Ref(MA(H - L, 22), -1) * 2 ,"ATR Mod",1.2,colorBlue);