Is adjusting your stop for a free/ reduced loss worth it from a probability standpoint?

[VEGASDESERT]

Its a loaded question, it comes down to knowing your system / strategy
inside and out.

Then you'll know when to move your stop loss or keep it in place or
move it further away for that matter..

 

[bone]

For swing trading, you don’t want tight stops. You don’t want daily modeled trading range volatility to take you out of a performing trade.

I have a great example. I have a client who was in a performing Corn Futures Butterfly (1-2-1). When he was marking up $200 he got cute and decided to put in a trailing stop. Sure as sugar he got taken out within a day.

Five trading days later the spread was worth $900.

So, for swing trading, we set our profit targets and our stop loss levels at the time of trade entry. If our indicator package flips before our stop-loss is reached, we are out of the trade (this happens for about half of our losses).

 

[speedo]

A good area to place a stop is where it makes technical sense. A bad reason to move it is to eliminate the fear of a loss in a trade. A pro trader learns to not fear losses or at least render the fear insignificant.

 

[lindq]

The only straightforward answer IMO is that moving your stop to B/E too soon is a guaranteed way of not making profits. Doesn't matter if you don't lose if you never make profits either.

On the other hand - it does not seem smart to let a winning trade turn into a loss either. So, I will typically B/E or more after some followthrough and rather re-enter if stopped (usually an option as I'm monitoring the trade in real time).

I will take one exception to that. There are instances where moving to break-even makes sense if the trade has really taken off, and a trader wants to protect the position against extreme market risk before the trade has run its course.

 

[ValeryN]

While @jharmon is absolutely correct, there is another way to put it.

Think of it this way - when you tested your strategy, you set precise rules. Let's assume those rules gave you edge and positive expectancy. If looking at what happens in the middle between entry and exit wasn't in your original test - then it doesn't matter. The moment you think you can make a decision based on anything else other than what you used when you tested - this is a new strategy. If you tested manually over 100 charts - then you need to go and run this new rule over those 100 charts and see what happens to this new system.

Think of any variation in rules creates a new system. It can be better, it can be worse. That is why you would hear someone saying - I tested million of strategies. They might have tested a single strategy from a discretionary point of view (let's say a MR on daily bars), but a million of variations in parameters values. 3 parameters with 100 values give you 1 mil outcomes.

When trading a system, sometimes we will see an extreme case. Where all our experience will be telling us - get out now. Or add. Or wait extra day etc. Whatever it is. And it might be something. In this case the best thing to do is to construct a simple test that simulate that exact situation, run it over a large data set to have enough samples, and see if doing what you think is right actually has a positive expectancy.

Val

 

[Laissez Faire]

I will take one exception to that. There are instances where moving to break-even makes sense if the trade has really taken off, and a trader wants to protect the position against extreme market risk before the trade has run its course.

Agreed.

And I believe that was what I alluded to when I said it doesn't make sense to let a winning trade turn into a loss?

 

[Bugsy]

If I am going for a swing trade (two legs or more) on successful Bull BO’s in the form of Strong spike my sl has to be just below the bottom of the BO because such BO’s can have deep PB’s after the first or second leg move and if my SL is too tight I will get stopped and just to see it then go in my favor. If I got in on the first leg up I would not tighten my SL until after the first PB or two ends and a resumption of the bull trend continues. I want to see higher lowers and higher highs after pb’s before I adjust my SL tighter.

However, if I am scalping 1 to 3 points say in the ES because I don’t think the BO is very strong I will scalp the BO several times compounding profits as it moves up. That is, scalp say 2 points..grab the profit let it drop a point or two enter again..then grab another point or two..ad nauseam all the way up on the spike till it spikes out. Often this will be several 1 to 3 point spikes on the same bar. Or it could be over a few bars.

A good area to place a stop is where it makes technical sense. A bad reason to move it is to eliminate the fear of a loss in a trade. A pro trader learns to not fear losses or at least render the fear insignificant.

While @jharmon is absolutely correct, there is another way to put it.

Think of it this way - when you tested your strategy, you set precise rules. Let's assume those rules gave you edge and positive expectancy. If looking at what happens in the middle between entry and exit wasn't in your original test - then it doesn't matter. The moment you think you can make a decision based on anything else other than what you used when you tested - this is a new strategy. If you tested manually over 100 charts - then you need to go and run this new rule over those 100 charts and see what happens to this new system.

Think of any variation in rules creates a new system. It can be better, it can be worse. That is why you would hear someone saying - I tested million of strategies. They might have tested a single strategy from a discretionary point of view (let's say a MR on daily bars), but a million of variations in parameters values. 3 parameters with 100 values give you 1 mil outcomes.

When trading a system, sometimes we will see an extreme case. Where all our experience will be telling us - get out now. Or add. Or wait extra day etc. Whatever it is. And it might be something. In this case the best thing to do is to construct a simple test that simulate that exact situation, run it over a large data set to have enough samples, and see if doing what you think is right actually has a positive expectancy.

Val

From my standpoint, and very similar to what Speedo stated, you can adjust it once doing so would mean a violation of the NEWLY adjusted stop would be a violation of your original premise when you made the trade. For example, I had initially planned to get in at a certain point as shown in the screenshot. It didn't dip low enough initially so I cancelled the trade and crashed out for the day. When I returned I saw it did in fact reach that entry point after I left and had since risen higher and was coming back again to the SAME entry point my system initially had it. I placed a limit order and it was filled before rising again. However, my initial stop price according to my initial play no longer made sense to me as a violation of the previous major higher low would in fact violate the premise of why I felt this trade would be successful. In doing so I was able to place my stop at the previous major higher low (original cancelled order) just under the initial entry I never actually entered into. This thereby reducing my risk while still being technically solid.

All in stating that had I entered the initial entry earlier it would be a reasonably good move to now move my stop under the low of that original, never entered into entry, as a violation at this point would in fact reverse the initial premise of this being a solid long play.