PacMan (my first hybrid) - page 40
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
What to do with runaway trades?
While rethinking about hedging again, basically our problem are "runaway trades" which fall out of our level hierarchy and are let loose until they return home.
If it doesn't return, we can only trade the other side to try to slowly compensate the floating loss until we can finally close out and restart.
Has anybody done some experiments concerning parallel trades also for the side currently having a loser? This would double our potential to compensate current loss, but may also increase risk of getting another runaway ...
Has anybody done some experiments concerning parallel trades also for the side currently having a loser? This would double our potential to compensate current loss, but may also increase risk of getting another runaway ...
I was thinking about this too and conclusion I made is that the only way to avoid losses (or to minimize chances of their occurance) is to have pipstep wide enough for the unexpected. Otherwise if say we leave runaway to crawl back and do some trading we must have enough margin to be able to avoid margincall. Bear in mind at this time one would have huge floating loss (runaway), and still must be able to afford another trade up to 5th or something progression. So capital requirement becomes so huge, that all this martingale thing looses any sense. Who would risk to bust his account any minute for 1 - 2% a month? Personally I would like to see something around 25% a month to play this game, otherwise risk / reward ratio becomes quite unattractive for me.
If we consider entries to be our main goal, what if we take some well - known EA and attach martingale engine to it? Say something from StepMA series, Phoenix or other? Those EAs don't have martingale to bail them out, so their entries must be quite good. Martingale would help them to deal with losing trades better, and we would get a better entry signal. Any ideas on that?
Only pacmanJr uses i_trend.
i need to change any setting with i_trend ?
eusd
ujpy
uchf
ucad
i need to change any setting with i_trend ?
eusd
ujpy
uchf
ucadno change, use defaults.
Dave
Personally I would like to see something around 25% a month to play this game, otherwise risk / reward ratio becomes quite unattractive for me.
Currently I am looking at about 20-25% gains per month running on 3 pairs. But this can vary. 15% one month and 40% another month,
But 20-25 average,
I would like to see 50% or more average per month. But that might be greedy.
dave
Currently I am looking at about 20-25% gains per month running on 3 pairs. But this can vary. 15% one month and 40% another month,
But 20-25 average,
I would like to see 50% or more average per month. But that might be greedy.
daveIs this with PacManJr?
...If we consider entries to be our main goal, what if we take some well - known EA and attach martingale engine to it? Say something from StepMA series, Phoenix or other? Those EAs don't have martingale to bail them out, so their entries must be quite good. Martingale would help them to deal with losing trades better, and we would get a better entry signal. Any ideas on that?
Yes, in principle this is a good idea, and actually this is what Dave has done in PacMan with his set of signals. From all my analysis and backed up by theory, we still need a signal with an edge for longer term success.
With random or near random signals and good settings one can still survive
a long time, but we are only moving bust further into future.
However, we also need to consider exit, which can make it more difficult to take over some strategy from a different EA. Note that with the martingale EA we never really want to close out a loser, because it hurts much more than in a different entry/exit strategy where regular losers may be part of it.
Sorry for repeating my (maybe stupid) question. 3 PacManJr on one account -no need to reduce Lotsize (risk)?
btw - is "Jr" the PacMan_X mentioned in an other thread?
Yes, in principle this is a good idea, and actually this is what Dave has done in PacMan with his set of signals. From all my analysis and backed up by theory, we still need a signal with an edge for longer term success.
With random or near random signals and good settings one can still survive
a long time, but we are only moving bust further into future.
However, we also need to consider exit, which can make it more difficult to take over some strategy from a different EA. Note that with the martingale EA we never really want to close out a loser, because it hurts much more than in a different entry/exit strategy where regular losers may be part of it.I agree - the better the entry signals the smaller the probability of a bust, but I'm sure that there is no strategy (other than time travel
) that will give us 100% success. So all we can do is move bust further into the future. Martingale users should also note that if they are unlucky BUST can happen on the first trade. This of course would be OK if it was possible to achieve > 100% profit / month. 25% is just not enough, because you will need at least 5 months of successful trading in order to compensate for ONE loss which will drain the margin account.
I still think that the best solution to this problem is to work out a well balanced exit strategy that will take some loss but will not drain the margin account.
I was thinking about closing portions of loosing trades with winners so that if a bust occurs it only drains some and not all of the margin account. If we are lucky enough we can still walk out with profit or break even.
I was wondering what you think about this approach. Has anyone experimented with something similar by manually exiting a runaway martingale progression?
...This of course would be OK if it was possible to achieve > 100% profit / month. 25% is just not enough, because you will need at least 5 months of successful trading in order to compensate for ONE loss which will drain the margin account.
You must bear in mind at what cost retuns are achieved with martingale strategies. To get more, you must get into position more often, risk more money, use smaller pipstep - so you must do everything that increases your chances to bust. Profitability here is a double edged sword, and if you increase your profit, mostly often you also increase your potential loss, so number of profitable months required to cover a loss either stays the same or even increases.
...I was thinking about closing portions of loosing trades with winners so that if a bust occurs it only drains some and not all of the margin account. If we are lucky enough we can still walk out with profit or break even. I was wondering what you think about this approach. Has anyone experimented with something similar by manually exiting a runaway martingale progression?
I tried to close losers manually for a couple of times, now I wish I had never touched them
My interventions made the whole situation even worse. Well, maybe I'm still not that good as a manual trader.
I came with one rough idea of loss handling: with the last increment we open another trade in opposite direction. Say we were buying eurusd, and it was going down. So we bought 0.1, 0.2, 0.4, 0.8 lots, and when buying 1.6, we also sell 1.6 lots. So we are 1.5 lots long in total. Now, if price starts to to up, we close 1.6 long with 1.6 short when long side hits tp, having lost anything but spread here, and hope price will go up even more to close the rest at 0.8 lots tp level. If price goes further down, we close our short side at next increments level with a few long orders starting with first (first 0.1, then, if we have enough profit, 0.2 or some of it etc.), and buy / sell 3.2 lots again. This stops loss from increasing exponentially (it increases linearly instead, but this is still better), and gives us more chances to wait for reversal. What do you think?