You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Exit methods for trend-traders
- Thread starter Cutten
- Start date
Quote from wwatson1:
Hi, I have picked the most recent trade in my journal to help show how I try and do things.
It is simple but I like it.
Would like to know if anyone else has the same kind of simplistic approach.
Here it is.........
This is good but it can be much better with minor improvements. Looks like you have a discipline in place so it should be a breeze for you to to put in one extra filter .
Quote from Cutten:
If the risk/reward is out of whack for a new entry, then it is also out of whack for someone currently in the market. The positions, after all, would be absolutely identical.
I'm not following.
Let's say the original entry long was at 100, with a five point stop.
Current price is 106.
Remembering that under this system the aim is to hold for a ten point gain, are you saying the odds are the same that the price will advance from 106 to 116 as from 106 to 110? What about when you factor in the stop? That is, the price could go back to 101, and while the original entry would still be long, wouldn't the new entry be stopped out?
Quote from spect8or:
I'm not following.
Let's say the original entry long was at 100, with a five point stop.
Current price is 106.
Remembering that under this system the aim is to hold for a ten point gain, are you saying the odds are the same that the price will advance from 106 to 116 as from 106 to 110? What about when you factor in the stop? That is, the price could go back to 101, and while the original entry would still be long, wouldn't the new entry be stopped out?
No. I'm saying that if the price is 106, then the odds of it advancing to 110 are identical whether you entered origninally at 100, or now at 106. The current risk/reward from the current market price is identical, regardless of where you entered.
It would be completely irrational for the "new entry" stop to be any different from the trailing stop on the already-open position. If it is correct to hold through a 5 point drawdown to 101 on an open position, then it is also correct on the same parameters/timescale to hold on a newly opened position during an identical 5 point drawdown to 101.
Interesting idea. However I disagree with the conclusion. Disagreement isn't a bad thing and if I am wrong I would like to find out and adjust my thinking.
Taking your example we assume the original long entry price is 100, the target is 110 and let's assume a stop loss of 95. For our purposes let's assume that we somehow know that this particular setup hits the profit target 40% of the time and the stop 60% of the time.
Now let's jump into the example where the setup has already occurred at 100, and price is now trading at 106. It is obvious that if we took an entry from 106 and rode it to the 110 target we would have 4 points of profit, and if it rode down to the 95 stop loss we would have 11 points of loss if we entered from this point and used the original stop and target points. This is clearly less than the 10 points of profit and the 5 points of loss dictated under our original risk/reward scenario. However I don't think this is what you are saying.
I think you're saying that at 100 we had a 40% chance to get to 110, and at 106, we still have a 40% chance to get to 110. I would dispute this point, citing the fact that the change in proximity to our target also unavoidably will change the likelihood of our reaching that point. For instance if we test entry every bar with a 5 point stop and a 10 point target, we will have a lower win percentage than if we test every bar with a 11 point stop and a 4 point target. The closer target is just more likely to get hit due to the normal fluctuations. You see, I don't believe in static probabilities in the market. I believe that the probabilities are constantly in motion. If we assume that we are now trading at 109.50 I'm sure you will agree that the probability of hitting 110 is much higher at that point than it was when price was 100.
I'm looking forward to hearing your arguments and clarifications as a very thoughtful member of this site.
Taking your example we assume the original long entry price is 100, the target is 110 and let's assume a stop loss of 95. For our purposes let's assume that we somehow know that this particular setup hits the profit target 40% of the time and the stop 60% of the time.
Now let's jump into the example where the setup has already occurred at 100, and price is now trading at 106. It is obvious that if we took an entry from 106 and rode it to the 110 target we would have 4 points of profit, and if it rode down to the 95 stop loss we would have 11 points of loss if we entered from this point and used the original stop and target points. This is clearly less than the 10 points of profit and the 5 points of loss dictated under our original risk/reward scenario. However I don't think this is what you are saying.
I think you're saying that at 100 we had a 40% chance to get to 110, and at 106, we still have a 40% chance to get to 110. I would dispute this point, citing the fact that the change in proximity to our target also unavoidably will change the likelihood of our reaching that point. For instance if we test entry every bar with a 5 point stop and a 10 point target, we will have a lower win percentage than if we test every bar with a 11 point stop and a 4 point target. The closer target is just more likely to get hit due to the normal fluctuations. You see, I don't believe in static probabilities in the market. I believe that the probabilities are constantly in motion. If we assume that we are now trading at 109.50 I'm sure you will agree that the probability of hitting 110 is much higher at that point than it was when price was 100.
I'm looking forward to hearing your arguments and clarifications as a very thoughtful member of this site.
Quote from Cutten:
No. I'm saying that if the price is 106, then the odds of it advancing to 110 are identical whether you entered origninally at 100, or now at 106. The current risk/reward from the current market price is identical, regardless of where you entered.
It would be completely irrational for the "new entry" stop to be any different from the trailing stop on the already-open position. If it is correct to hold through a 5 point drawdown to 101 on an open position, then it is also correct on the same parameters/timescale to hold on a newly opened position during an identical 5 point drawdown to 101.
when u get the exit signal looking the same as the entry signal u r there. if u r trading 2 or 3 min charts u have got hell to pay anyway..forget it just jump in when u can't stand it and jump out when u can't stand it...scalping is only for the best traders not the weakest as I am......if u r trading a short chart try to look at a longer one such as 15 or 30 for clarity of run/exit, etc. zoom out not in......
Quote from MaxMin:
Interesting idea. However I disagree with the conclusion. Disagreement isn't a bad thing and if I am wrong I would like to find out and adjust my thinking.
Taking your example we assume the original long entry price is 100, the target is 110 and let's assume a stop loss of 95. For our purposes let's assume that we somehow know that this particular setup hits the profit target 40% of the time and the stop 60% of the time.
Now let's jump into the example where the setup has already occurred at 100, and price is now trading at 106. It is obvious that if we took an entry from 106 and rode it to the 110 target we would have 4 points of profit, and if it rode down to the 95 stop loss we would have 11 points of loss if we entered from this point and used the original stop and target points. This is clearly less than the 10 points of profit and the 5 points of loss dictated under our original risk/reward scenario. However I don't think this is what you are saying.
I think you're saying that at 100 we had a 40% chance to get to 110, and at 106, we still have a 40% chance to get to 110. I would dispute this point, citing the fact that the change in proximity to our target also unavoidably will change the likelihood of our reaching that point. For instance if we test entry every bar with a 5 point stop and a 10 point target, we will have a lower win percentage than if we test every bar with a 11 point stop and a 4 point target. The closer target is just more likely to get hit due to the normal fluctuations. You see, I don't believe in static probabilities in the market. I believe that the probabilities are constantly in motion. If we assume that we are now trading at 109.50 I'm sure you will agree that the probability of hitting 110 is much higher at that point than it was when price was 100.
I'm looking forward to hearing your arguments and clarifications as a very thoughtful member of this site.
I think attaining the 110 even when at 109.50 depends on how extended the trend is, support and resistance levels, avg daily range, standard deviation and other volatility factors. Depending on the latter it might have a less probability of attaining a profit target as you move closer to it.
Outside of news and a major shift in valuations, one might exit portions of the position a little bit sooner than previously targeted. The changing of probabilities sets of continuous market motion.
I suspect that it does not matter exactly what trailing stop you use. The markets are messy enough that lots of things work about as well as lots of other rules.
Quote from MaxMin:
I think you're saying that at 100 we had a 40% chance to get to 110, and at 106, we still have a 40% chance to get to 110. I would dispute this point, citing the fact that the change in proximity to our target also unavoidably will change the likelihood of our reaching that point. For instance if we test entry every bar with a 5 point stop and a 10 point target, we will have a lower win percentage than if we test every bar with a 11 point stop and a 4 point target. The closer target is just more likely to get hit due to the normal fluctuations. You see, I don't believe in static probabilities in the market. I believe that the probabilities are constantly in motion. If we assume that we are now trading at 109.50 I'm sure you will agree that the probability of hitting 110 is much higher at that point than it was when price was 100.
I'm looking forward to hearing your arguments and clarifications as a very thoughtful member of this site.
No, you've misunderstood my point I think. I agree that at 106 that you are more likely to hit your 110 target, for the reason you give i.e. that the market is closer to it. All I'm saying is that the risk/reward at 106 has nothing to do with whether you are flat or already long. Either being long at 106 is a good position i.e. the odds favour profit, or it is a neutral or bad position. These odds have nothing to do with whether you have already managed to capture the move up to 106 and are still long, or whether you sat it out and are currently flat.
My basic point is that your current position, and any open profit, has no effect at all on the current risk/reward offered by taking a position in the market now.
RE: "Hi, I have picked the most recent trade in my journal to help show how I try and do things.
It is simple but I like it.
Would like to know if anyone else has the same kind of simplistic approach.
Here it is........."
wwatson1: It seemed ot me that your graph uses a simple scale for the price axis. May I suggest using a log-scale (for the kind of time periods you seem to trade). I move from 50 to 100 is more like a move from 100 to 200 (not to 150).
Trendlines (etc) could look quite different on a log-scaled graph.
It is simple but I like it.
Would like to know if anyone else has the same kind of simplistic approach.
Here it is........."
wwatson1: It seemed ot me that your graph uses a simple scale for the price axis. May I suggest using a log-scale (for the kind of time periods you seem to trade). I move from 50 to 100 is more like a move from 100 to 200 (not to 150).
Trendlines (etc) could look quite different on a log-scaled graph.