Set it and Forget it, Dollar Cost Averaging | Page 8 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Wide stop loss, Dollar cost averaging or doing nothing 18 replies

Best Indicator For Dollar Cost Averaging Into Crypto 0 replies

10,000 MA Dollar Cost Averaging 6 replies

The Dollar Cost Averaging Pivot Point Trading Method 914 replies

Fast Averaging versus Slow Averaging 1 reply

Attachments: Set it and Forget it, Dollar Cost Averaging
Exit Attachments

Set it and Forget it, Dollar Cost Averaging

  • #141
  • Aug 22, 2012 9:19am Aug 22, 2012 9:19am
Quoting jaws810
I've been looking at posts and trying for solve the issue with huge runs and massive DD and here is what I came up with.

I think I found a way to not only survive, but thrive through a hellish death spiral. Trade in both directions, use multiple limit orders (200+ pips out) to catch big moves and use a 150 pip from average Sl to close all positions and limit losses during a big move. Then when you get a huge move like that September 2008 you make a killing on the winning side and losses are minimal as long as you set up enough limit orders to...
Jaws,
I don't want to sound rude, but I think you are just throwing darts w/o really thinking about what you are trying to accomplish. And, you need to backtest, whether it's visual or rigorous...

Before throwing away your original idea, you should think about the market, market moves, ranges, and how to tweak your idea. Like others have suggested, the EURUSD is not a prime candidate for a reversion to the mean (RTM) strategy, which is what you are ultimately using. I have downloaded ticks for AUDUSD, USDCAD, USDCHF to see if they are candidates. At least start with the right market.

Then and only then can you start looking at x pips this way, y pips that way. Otherwise, it's just guestimating...

Just my humble two cents...
BroncoCap
  • #142
  • Aug 22, 2012 9:28am Aug 22, 2012 9:28am
Quoting BroncoCap
Good analysis, but note. My results were Jaws' original system. Open only 1 trade per day at a given time. Therefore, the system I tested is a likely loser because losses (aka many positions) form an exponential negative equity curve whereas gains are linear. That is to say, winners only give a fixed amount (x pips) whereas losses are unlimited.

Said another way, think of it like this. In any given week, there will be 10 total trades, 5 long and 5 short. The absolute max gain would be both sides hitting targets and accts flat over the weekend...

I totally agree with what you're saying about the exponential vs. linear. Now Jaws did mention is his first post I believe about adding trades if priced moved interday by 50 pips. But I do think we need to convert the profit side to exponential by not closing them immediately in order to make it work.
  • #143
  • Aug 22, 2012 9:29am Aug 22, 2012 9:29am
Quoting spyderman
So Bronco has run several tests and mentions the huge amount of positions opened before a retracement closed them. I don't think that is really the issue. Those many positions should have been offset by the many positions that would have closed at t/p.
Once again, that's not true.
Suppose you enter a buy and a sell at price X. For the sake of the example, let's say the price moved 100 pips up that day and you use 100 pips TP. So, on the next day you have your buy hit TP, then you should enter another buy/sell pair, am I right? A buy to cover the upside and a sell to average the losing sell from yesterday, right? Now, suppose the price moves another 100 pips up. Another TP and another buy/sell pair, for the same reasons. However, at that moment you'd have +200 pips from take profits and -300 pips floating loss from sells. Another 100 pips up and its +300 vs -600. It might look like it has some effect, but it's only initially. Sequences that have DD problems with averaging will have the exact same problems regardless of taking both sides or not. Because the deep DD is caused by accumulating losses that grow exponentially and the profits are only linear. At some point, the profits collected just become irrelevant. You have 10x or 20x your normal position size, on the wrong side of the trend. It's like walking backwards on the railway, hoping that the extra distance will allow the train to stop before hitting you. Personally, I'd get off the damn track.

Best regards,
Barbossa
  • #144
  • Aug 22, 2012 9:34am Aug 22, 2012 9:34am
Here are the 2010 and 2011 charts for USDCAD...something like this could work much, much better:

Attached Image (click to enlarge)
Click to Enlarge Name: USDCAD.2010.jpg Size: 200 KB


Attached Image (click to enlarge)
Click to Enlarge Name: USDCAD.2011.jpg Size: 181 KB
  • #145
  • Aug 22, 2012 9:37am Aug 22, 2012 9:37am
Quoting BroncoCap
Jaws,
I don't want to sound rude, but I think you are just throwing darts w/o really thinking about what you are trying to accomplish. And, you need to backtest, whether it's visual or rigorous...

Before throwing away your original idea, you should think about the market, market moves, ranges, and how to tweak your idea. Like others have suggested, the EURUSD is not a prime candidate for a reversion to the mean (RTM) strategy, which is what you are ultimately using. I have downloaded ticks for AUDUSD, USDCAD, USDCHF to see if they are...
I know that the euro might not be the best for this system but if I get it to work with the euro it should work with most pairs.
  • #146
  • Aug 22, 2012 9:46am Aug 22, 2012 9:46am
Hello JAWS 810!

Thanks for this interesting thread so far, I have couple of questions in my mind, I do not know if I could make
Myself clear, here are the questions:-

1) Whenever TP hits either side(SHORT/LONG), do I have to take another position on BOTH sides??


2) At least(minimum) how many position(Short/Long), will remain OPEN all the time on both sides.

Thanks,

Regards.
Sevenskies.
  • #147
  • Aug 22, 2012 1:28pm Aug 22, 2012 1:28pm
in my system I dont take every trade.

you got to remember that the profit is coming from buying low and selling high.

once I have 10 trades on say the buy side I get very picky for my next trades. for my 11th trade I only take trades which are at least 50 pips down from yesterday`s close. the 12th trade 50 pips down again.

by trade 13 I go down 50 pips again + I close every trade that is in profit, regardless of the amount of pips on the buy side.
  • #148
  • Aug 22, 2012 1:30pm Aug 22, 2012 1:30pm
Quoting Barbossa
Once again, that's not true.
You're right about the linear vs. exponential. That's why I mentioned in another post that you would have to figure a way to leave your buys (in this case) open to accrue exponential profits.

So in your example. If rather then TP at 100 you just added positions every 50 pips. First day 1 sell 2 buys, second day 1 sell 2 buys. At end of day 2 you're -300 on the sells and +500 on the buys.

I'm still trying to figure out the MM and how to close at the best time but at least we're positive pips at this point.

Your thoughts?
  • #149
  • Aug 22, 2012 1:35pm Aug 22, 2012 1:35pm
Quoting jaws810
I know that the euro might not be the best for this system but if I get it to work with the euro it should work with most pairs.
????????????

are you trying to prove something or do you want to make money
  • #150
  • Aug 22, 2012 2:13pm Aug 22, 2012 2:13pm
Quoting spyderman
You're right about the linear vs. exponential. That's why I mentioned in another post that you would have to figure a way to leave your buys (in this case) open to accrue exponential profits.

So in your example. If rather then TP at 100 you just added positions every 50 pips. First day 1 sell 2 buys, second day 1 sell 2 buys. At end of day 2 you're -300 on the sells and +500 on the buys.

I'm still trying to figure out the MM and how to close at the best time but at least we're positive pips at this point.

Your thoughts?
Close when the whole bundle reaches predefined profit could be fine.
By the way, the idea of building exponential winners instead of losers sounds more sensible to me. However, you'll need a backup plan in case of whipsaws. I like getting more positions on the winning side, but you have to make sure it doesn't go against you.

Sincerely,
Barbossa
  • #151
  • Aug 22, 2012 2:15pm Aug 22, 2012 2:15pm
Opening a Buy and Sell = a single directional order at some point.
The moment you close one side of the Buy and sell, you have placed
an order for the other side. Why bother doing it ?
You may as well place the single order and save the spread.
  • #152
  • Aug 22, 2012 2:22pm Aug 22, 2012 2:22pm
Quoting Barbossa
Close when the whole bundle reaches predefined profit could be fine.
By the way, the idea of building exponential winners instead of losers sounds more sensible to me. However, you'll need a backup plan in case of whipsaws. I like getting more positions on the winning side, but you have to make sure it doesn't go against you.

Sincerely,
Barbossa
If I have crystal ball and know the winning side, I put all money in.
Follow me to the Promised Land
  • #153
  • Aug 22, 2012 2:23pm Aug 22, 2012 2:23pm
Quoting Barbossa
Close when the whole bundle reaches predefined profit could be fine.
By the way, the idea of building exponential winners instead of losers sounds more sensible to me. However, you'll need a backup plan in case of whipsaws. I like getting more positions on the winning side, but you have to make sure it doesn't go against you.

Sincerely,
Barbossa

Yeah its kinda like basket trading. Close when the basket looks good. Whipsaws are always tough. Course if they're real tight things will just flatten. But if you were getting a constant 75 pip back and forth that might be tough. Hard to imagine all the possible scenarios. Maybe I'll see if I can get an EA put together over at Steve Hopwoods to do some testing.
  • #154
  • Aug 22, 2012 6:36pm Aug 22, 2012 6:36pm
There's been some discussion about pair selection and whether or not E/U is a good candidate.

To my way of thinking the way to profitability is larger moves in a given direction allowing additional orders (additional to the standard buy and sell of the day) to be placed and grow exponentially. So to that end, I would think E/U would be an excellent candidate. It's been one of the larger movers of late.
  • #155
  • Aug 22, 2012 6:44pm Aug 22, 2012 6:44pm
Quoting spyderman
There's been some discussion about pair selection and whether or not E/U is a good candidate.

To my way of thinking the way to profitability is larger moves in a given direction allowing additional orders (additional to the standard buy and sell of the day) to be placed and grow exponentially. So to that end, I would think E/U would be an excellent candidate. It's been one of the larger movers of late.
You want a market that is trading in say a 1000pip range or less, it's enough to take profits but not enough to go to far and bankrupt you.

Ofcourse there is always a risk of going and going.

EU is in a daily downtrend, it has down months with barely any up moves so TRAPPED!!

GU is better, 1000 pip range no real trend over the last year, fine.

EJ is to straight liney.

AU same.

Haven't got any other pairs up, but it looks like GU or nothing.

Not ideal market conditions for this really.
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #156
  • Aug 22, 2012 6:57pm Aug 22, 2012 6:57pm
Quoting Turveyd

EU is in a daily downtrend, it has down months with barely any up moves so TRAPPED!!

But again, why would that be a bad thing. If price moved steadily down, down, down, your daily buy orders (in drawdown) would be offset by your daily sell orders. Meanwhile, you'd be making your profits on the extra interday sells, no?
  • #157
  • Aug 22, 2012 8:30pm Aug 22, 2012 8:30pm
Quoting spyderman
But again, why would that be a bad thing. If price moved steadily down, down, down, your daily buy orders (in drawdown) would be offset by your daily sell orders. Meanwhile, you'd be making your profits on the extra interday sells, no?
No, because you would only be making 50 pips per day on the winning side, but on the losing side, EVERY position is drowning further another 50 pips .
  • #158
  • Aug 22, 2012 8:33pm Aug 22, 2012 8:33pm
Quoting SmartGXL
No, because you would only be making 50 pips per day on the winning side, but on the losing side, EVERY position is drowning further another 50 pips .

No, because I'm not closing my buys at 50 pips, I'm leaving them open
  • #159
  • Aug 22, 2012 8:46pm Aug 22, 2012 8:46pm
Quoting spyderman
No, because I'm not closing my buys at 50 pips, I'm leaving them open
I agree we need the most trending pair. You have the best idea so far, but what if it takes pending buy order and comes down and takes pending sell and goes and for a while, may be days.....how to decide at what intervals we need to put the pending orders 50, 100 pips for the best results?
If the basket goes positive at what point you will decide to take profit 100, 150, 200 pips?
  • #160
  • Aug 22, 2012 9:25pm Aug 22, 2012 9:25pm
Quoting jeuro
At 5% per week... those 5k would produce 250 per week ... that would be some 2 years just to breakeven...but wait.. what about if that happens again in a couple months??


J.
250 per week gets to 5 k in just 20 weeks. However 5% compounding growth takes about 15 weeks for 5K to get back to 10k
Kill the ego, or it will kill the mission.
0 traders viewing now