Stoploss.......Theory of Failure.

[ducati998]

KIWI,

Yes, I do obviously have an alternative. And it will be unleashed on the site by "FRUGI" as and when.

ROGUE,

Risk, is present within the financial markets, and the management of that risk requires an underlying methodology to manage the risk.

The "STOPLOSS" is an inappropriate tool. It was adopted from professional gambling, and bolted onto the financial markets.

In gambling it may well be appropriate. For traders it is one of the weak links.

FRUGI

Losses are necessary, unavoidable costs entailed in the business of trading, much as staff costs are for a supermarket giant. The profit is not "provided" by the cutting of losses: losses are cut to ensure that the profitable trades are not significantly eroded.

To an extent losses, or the potential for loss will always be present within the financial markets.
However that is a very different proposition from equating them with a cost of doing business.
An example of a cost of doing business is one of brokerage costs.
A stoploss is NOT a cost of doing business.
It is a "COST" on profit margins........there is a very real difference.

In the same way the cutting of staff costs does not "provide" the profit for the supermarket giant, as the profit comes from selling products. A profitable business that ruthlessly cuts its costs tends to do better than one that doesn't, surely?

In the example "employee" costs to the supermarket are indeed a cost, and must be managed.
If they are not managed within the available "profit margin" then profits will shrink or disappear.
And of course as business and technology progress new and improved methods are implemented to control or cut costs, to improve profitability. Look at "Online" supermarket shopping.........it may never fully take off, but it is an experiment to minimise staffing costs in this industry. So in answer, in the supermarket business, costs of employees and cost of product, are interlinked and calculated to provide the profit margin in the final consumer price paid. ( with all other costs omitted in this discussion ) And until employees can be eliminated, they are a cost of doing business.

Any business enterprise that relies on the ruthless cutting of losses to provide a 'profit' has serious flaws

Hence my quote, stoplosses are not the control of costs, they are the control of losses.
And yet, if said trader were to trade on a purely "technical" basis, without utilising stoplosses, almost certainly they would be wiped out.

That is why the question...................
If "Stoplosses" impact profitability in a negative way, and in point of fact contribute directly to most traders losses, especially in the early days;
Yet are vital to the methodology,
Is not the methodology seriously flawed?

StkWinGuy,

First, a stop loss is an insurance tool to assure that you will conserve enough capital to be around when the right situation does comes along - there is no quarantee WHEN that will occur.

Interesting.
Stoplosses are not an "insurance" tool, they are a gambling tool.
They ( stops ) very gradually erode your capital.
They do so as your methodology is so weak, that the stop becomes necessary to keep you in the game ( your methodology as in "TECHNICALS", not you personally )

Right situation..............TIME, yes agreed here, time is a whole new topic however.

cheers d998

 

[roguetrader]

Ducati, you still have not defined what a stop loss is, by your definition. I have provided one such definition I would be interested to see how far they are apart.

 

[RUDEBOY]

Ducatti, are you saying that stops cause loss sometimes?

 

[roguetrader]

They ( stops ) very gradually erode your capital.

Sorry but that is a ridiculous statement.
Stoplosses by any definition, will only erode your capital if they are coupled with poor trade selection and decisions.
For someone who has made north of $12,000 a month using TA and stoplosses, I find your positioning somewhat of a, as Greenie would say, conundrum

 

[RUDEBOY]

RT, this is very true. Trading is about intergrated issues, not seperate issues!

 

[RUDEBOY]

Is there anybody on here a person would trust to make thier trades for them?

 

[RUDEBOY]

The only person i would trust is.............? What would you think of a person who was in control of your account? Rude!

 

[nine]

Rudeboy,

It is absolutely true that stoplosses cause losses some of the time ... in fact they cause loss much of the time.

Stridsmann does a few studies of adding stop losses to systems. In all cases they reduce the overall profitability of the systems (we all know how often price comes back just after hitting the stop). His arguements for them are two fold:
- to provide a defined risk that is used for money management/risk control; and
- to reduce the variability of your systems results (std deviation)

Both of these points improve your potential return when using agressive money management.

 

[roguetrader]

It is absolutely true that stoplosses cause losses some of the time ... in fact they cause loss much of the time.

Hi Kiw, could you elaborate perhaps on how these stoplosses go about causing a loss?

 

[frugi]

D998,

So in answer, in the supermarket business, costs of employees and cost of product, are interlinked and calculated to provide the profit margin in the final consumer price paid. ( with all other costs omitted in this discussion ) And until employees can be eliminated, they are a cost of doing business.

And therein lies a fine parallel with trading losses They too can be treated as simple costs and included in the profit margin calculations. Neiher will a supermarket sell product without any staff; nor will a trader will generate consistent profits without accepting losses. The only difference between broker commision / fixed staff costs etc. and losses, in terms of overall profit, is that the trader decides on the latter and thus flexibly affects the profit from trade to trade..

That is why the question...................
If "Stoplosses" impact profitability in a negative way, and in point of fact contribute directly to most traders losses, especially in the early days;
Yet are vital to the methodology,
Is not the methodology seriously flawed?

Stop losses contribute directly to most traders losses

Of course, as that is their function. In fact, they don't merely contribute: they provide every loss - and we are glad of them because they are simple and easy to take. It's the open profit that kills.

Stop losses "impact profitability in a negative way

Microscopically, of course, each loss will erode open profit. Each single wage packet erodes open profit. No problem with that. However, taking each and every loss at an appropriate time can ensure that, over time, the overall profit is left largely intact. Profits cannot consistently be taken unless one accepts the probability of small losses along the way, at least as far as the naive methods I employ are concerned. I treat poker in much the same way and am glad of the professional gambling principles that are transferable between the disciplines.

I don't understand why you feel the methodology is flawed?

A stoploss is NOT a cost of doing business.
It is a "COST" on profit margins........there is a very real difference

Dare I mention semantics again?

I am dispossessed of any accountancy skills so am unable to distinguish between a cost of doing business and a cost on profit margins; at least, in this discussion, I view them as much the same since in either case the profit is kept separate. The cost of doing business will transfer a cost onto the profit margin, as it were, so why the need to distinguish between them? Of course there may be a real difference on ledgered paper but the important point is to ensure that the total stop losses taken only form a small percentahge of the total profits.

Your first post was thought provoking and your article-to-be too.

 

[nine]

Yes (and I do detect the meaning in your bolded text ) but to get real elaboration you have to read stridsmanns first book on system development.

The best way of mentally illustrating how stop losses cause loss is to think about stop losses that are too tight. In this case the normal chop of the markets hits the stop and their presence causes loss.

What stridsmann illustrates is that just about any level of stop loss point causes a reduction in system profitability by exiting on a number of swings that then return - that the overall losses on those that get exited from and return before a reversal signal occurs exceeds the reduction in losses on the ones that get exited from before you get a reversal signal.

Now I'm confused. He spends a couple of chapters on it and then explains why its still a good idea to use them. The book is worth reading.

But the point is that, at almost any level of tightness, for most systems (I am sure there will be exceptions) then stop losses "cause" systematic reduction in profitability on a fixed risk trade basis (not taking into account the increased agressiveness that they allow by reducing std deviation and defining the loss).

My concern about ducati's thread (and hence my question in the 2nd post about his alternative, not yet proposed) is that he may be taking the simplistic view of stop losses where they cause loss. This view is true .... but ... it may not take into account variability, money management and risk of ruin.

 

[roguetrader]

Thanx for replying Kiwi, you have interpreted the embolding of the text exactly as it was meant to be interpreted.

The best way of mentally illustrating how stop losses cause loss is to think about stop losses that are too tight. In this case the normal chop of the markets hits the stop and their presence causes loss.

This loss was caused by the trader innappropriately placing the stoploss. I only make that point to give you some idea of where I amcoming fromon the issue of stoplosses, i realize this is not the crux of your view on them.
Other losses as a result of , as you pointed out price moving down to hit a stoploss only to retrace and move on, are also caused by poor placement, not the stoploss itself, such stoplosses are in predictable locations.

The reat of your answer suggests to me that there are things which I have possibly not considered or taken into account, as it pre-answers some of my suggestions so I would not be prepared to debate further from this perspective without being fully familiar with the Stridsman subject matter. You have me at a dis-advantage.

I do not use conventional stoplosses so I do not defend them from a point of bias merely my observations on the subject.

 

[nine]

Yes

I would be as interested in your approach to stoplossless trading as I am in ducatis.

I note that I recently raised my standard stop on something to 5 points past the swing point (it was 3) because giving up the extra 2pts on some occasions was far outweighed by giving up the range from entry to the prior stop loss on a very few occasions.

 

[dollar]

Its a matter of perception. Its the word "LOSS" which gives negative connotation to the term 'stoploss'.
I'd like to treat stops as a business expense like insurance premium - undesirable but necessary.
Just my 2p , ofcourse.

 

[Trader333]

All the traders I know who didnt use stoplosses are now Ex traders with insufficient funds to be able to trade with. Personally I would rather use stoplosses and be able to trade another day.


Paul

 

[JumpOff]

I can assure you that pigs do not fly.

I am very sorry to hear this!
JO

 

[barjon]

Yes

I would be as interested in your approach to stoplossless trading as I am in ducatis.

I note that I recently raised my standard stop on something to 5 points past the swing point (it was 3) because giving up the extra 2pts on some occasions was far outweighed by giving up the range from entry to the prior stop loss on a very few occasions.

Kiwi

Interesting and your reference to Stridsmann.

Kaufman, also, has some good thoughts on stoplosses. Firstly, he shows that the size of stops results in losses very close to a constant value:

Size of stop...................5.................10..................20
No occurences............20...............10..................5
Loss..............................100.............100...............100

Secondly, he conducted a 10 year test of stoplosses (from .02% to 10%) applied to a simple moving average trend following system across four markets with trend periods ranging from 5 to 300 days. Overall, the performance of the trend program without a stop loss proved to be best for the range of trend periods, although specific stops were better for some intervals.

Despite these results, he says, it is difficult to trade without a clear idea of risk. Stops help here but a system should always be tested to compare with performance that does not use a stop. The safest way to reduce risk, he concludes, is simply by reducing the size of your positions.

ps: nb, the analysis was about fixed stops and not those based on support/resistance which may be quite different.

good trading

jon

 

[peterpr]

Its a matter of perception. Its the word "LOSS" which gives negative connotation to the term 'stoploss'.
I'd like to treat stops as a business expense like insurance premium - undesirable but necessary.
Just my 2p , ofcourse.

Best point so far in what was no doubt intended as a provocative thread opening.

At further risk of 'semantics' accusations, I don't use 'stop-losses' either. I use stops - and limits and several other order types to manage a trade. I enter a trade based on a probability assessment; I set stops to deal with the inevitable probability exceptions and limits to close when the original assessment is confirmed.

I have no problem with the professional gambling analogy either - unless it is intended to convey some kind of value judgement - subject for a different discussion perhaps?

 

[Glenn]

Any business enterprise that relies on the ruthless cutting of losses to provide a "profit" has serious flaws.

There seems to be an element of "a bad workman always blames his tools" here.

e.g. If you don't know how to wield a hammer, then don't blame the hammer when it hits your thumb.

Trading is inherently risky. The risk has to be managed in order to succeed.
When you enter a trade, anything can happen no matter what you may think, or why.
Once you have placed your trade, you are entirely at the mercy of all the other traders of that instrument as to its direction thereafter.
So for every trade you have to define how much you are prepared to risk (lose) in order to find out if you are right.
The amount you are prepared to lose is defined by your initial stop - actual or mental.
If you are not adept at choosing this then you may be stopped out unnecessarily, resulting in an unnecessary loss.
But it was not the stop itself which caused that loss, it was the decision made by the trader on where the stop should be which caused that loss.

Glenn

 

[roguetrader]

Secondly, he conducted a 10 year test of stoplosses (from .02% to 10%) applied to a simple moving average trend following system across four markets with trend periods ranging from 5 to 300 days. Overall, the performance of the trend program without a stop loss proved to be best for the range of trend periods, although specific stops were better for some intervals.

Hi barjon, this concept of no stoploss of any kind appears to hinge on the principle that a corectly analysed and selected trade will most times exhibit a profit at some point in its life, if you hold it for long enough, however what the model would fail to show if this were the case is what use the capital tied up in the trade could have been put to in the meantime.