Apologize for the long delay.
Whatever you have written is correct:
As price makes one sided movement the drawdown is huge because there are a lot of open opposite orders.
Swap could kill the account.
Manual intervention is important:
For eg the Lehmann brothers news in 2009. We could have just started placing primers after the news came out with 100pip TP and new primers every 30 pips with auto SL as per our rule.
You could also let the EA place normal trades as it does.
This would have saved the channel and made enormous profits during the huge drop.
What is important lessons in this strategy is mathematical formulas of:
Normal trades
Primer rules
All we need is to implement both these strategies properly.
By avoiding greed and with due diligence we can expand the channel and let it run profitably within 2 years.
With the above explanation I still think that my first logic of using ChannelBottom and ChannelTop would be useful to trade the channels. All it would need is for the user to adjust the values of ChannelTop and ChannelBottom with a bit of market research.
Lets take an example (this is purely my analysis):
AUDUSD (because i prefer this pair as it has been ranging for quite a long time:
Whenever there is a rate cut - You set the channelbottom at the price rate cut has happened and let primers take over.
Whenever rate is held stable or increased you fix channeltop values to let buy primers take over.
You could use the above based on important data like Unemployment or China trade etc.
I am planning to start an AUDUSD channel next month and am going to use my ChannelTop/ChanneBottom EA to trade it. I feel trying to figure out the perfect mathematical way to use this EA might save a channel but to use it effectively and build consistent profits we cannot shirk our responsibility of following the pair and using market data. The important things to learn from the Greenland strategy are the normal trade rules, primer rules, primer sizes - which are already implemented in the EA.
I think ChannelTop/Bottom should be set automatically when primers are triggered, not manually due to human interpretations. The drawdown calculations are quite frankly only another way of saying P&L. Primers main function is to neutralize negative P&L by increasing opposing trades, so order sizes should be calculated from account balance, not equity, and primer sizes from opposing open trades needed to be neutralized. MA has no place in this logic, instead we need fuzzy logic for the primers placement.
A static 30 pip primer placement does not work as regular retracements are too big too often. Having larger primers where only one is unsecured and extending TP rather than always placing new primers is needed. Closing many small primers increases account balance while decreasing equity, especially when SL is hit. This gives an add-on effect of increasing regular trade size too often, and close to the borders too, which again will decrease equity.
What we need are longer primers that better support P&L so that equity moves back above 40% for as long as possible. We can calculate the drawdown, and then get how many pips an opposing trade of size x can cover the losing trades until an equity limit is hit again. At that limit, another primer is needed in addition (not instead of P1). Only this way will we be able to extend the channel “indefinatly”. Smaller primers/sub primers etc. would be nice for increased profits, but I don’t see that happening without manual interpretations of each situation. The problem now is of course that we don’t have these calculations yet.
CC,
Im with Dennis on this one, ChannelTop/Bottom should be set automatically to prevent human error.
Dennis, just to clarify, what are you thinking with the primers? making all primers with a fixed gap size, just a big one (like 80 pips) or did you think of also trying to calculate the gab size on the fly for each primer run?
Calculate everything on the fly. Since I don’t have the calculations yet I can’t be exact, but I think the width of the channel and the area covered by increased lots will change how primers should be placed.
I note your points. I agree that to avoid human error and to have a fully functional EA primer functionality should be automatic. The last version of the EA places primers based on drawdowns but to see the trend direction it makes use of the MA. Please suggest another method of finding out the trend and i will try to implement it.
I just read Viking’s interview where he says that setting a 30 pip Primer Step is too small for the cable and that it should have been 80 pips bcos the retracements on cable is large.
Right now we only carry 1 primer if rate turns, right?
I will try to do some testing myself, but based on the results that have been presented here so far there are still some issue with getting the primers to function properly?
At the start of the channel (lets take cable for eg) we fix the channeltop and channelbottom values based on our calculation of balance.
Lets say we start at 1.5800 with $2k. and we can survive 400 pips on either direction.
Hence,
ChannelTop = 1.6200
ChannelBottom = 1.5400
Primers will be started to place once price breaks either of these values.
ChannelTop & Bottom values will follow price but be below or above the current price by userdefined step.
Eg. Price moves below 1.5400 - primers will be placed. Lets say user defined step for AutochannelValues is 300 pips,
then once price reaches 1.5099 ChannelBottom value will become 1.5399 and continue to go down as prie moves down. This will also expand the channel and normal trades will be placed once price goes back above channelbottom.
Primerstep & PrimerTP for individual pair has ot be identified suitably.
Is merely having a channel top/bottom sufficient? Shouldn’t the ‘drawdown’ come into place as well?
Say the price stays within the channel for 4 months, allowing you to build up your capital and equity. That increased capital should make the channel expand, instead of just automatically introduce primers?
I thought the primers were placed based on the 0.6% DD rule?
This will not do. Think of primers in the context of profits and loss (P&L = equity - account balance). You have a certain amount of open orders that are putting increasingly large negative pressure on P&L as rate moves against them bringing down equity. P&L is the net worth of all your open orders, positive and negative, therefore, if you have a lot of longs in minus, opening a short of the same size as the combined longs, your P&L will suddenly return to 0 (minus spread). If rate goes down the longs will decrease P&L, but the short primer will increase it by the same amount, neutralizing P&L and hence equity, right? (And also reducing used margin depending on your brokers rules for hedging).
This has nothing to do with distance per se, so setting channelTop/Bottom at startup is meaningless. You can have orders of different sizes and at different places within your natural grid of regular trades. You also cannot know beforehand how much you make on ranging before moving outside those original +/- 400 pips. However, at some point when P&L becomes a problem to cover using only your current account balance, that’s when a primer kicks in. Hence, opening rate for the primer is where you set the channel’s current border in order to stop placement of regular trades below/above. The point of primers is to maintain this balance of equity, not “making more money” although that is a nice side effect of course
When a primers is taken out by SL, the channel border has to be recalculated, but I don’t have good info on that part yet. If P1 closes at +10 it’s dangerous to move the border to open rate for P2, especially if its 80-100 pips away. I’m still working on this part, and how to calculate what TP/new primer distance to use in each situation. It would be great if someone can get Michael to answer a few more questions about this aspect…
Agree with what you say Dennis. That is what we were trying to do with the previous versions of the EA.
When you come out with a proper logic I will try to implement again.
I will be actively involved in this thread. But I have decided to start a channel on AUD USD myself from next month. I am going to use manual trades also as Michael is doing. I am pretty confident on using the primers too (hopefully not over confidence).
To reduce swap losses I am planning to use two brokers, one for buys and one for sells. I will let the EA place normal trades and jump with manual trades by following news and market sentiment.
The only thing I hope is to not become greedy which i have often become previously
You realize you’re going to need a lot more money for that without saving any swap don’t you?
As long as buys/sells are balanced the swap will somewhat even out (earn swap on one, pay for the other). Furthermore, used margin (the money taken out of you account, or “frozen” when opening an order) will increase. Depending on your broker, your used margin is either halved or 0 for equal buy and sells on the same symbol. With two brokers (or even 2 accounts at same broker) you’ll need to cover full margin for all trades, all the time. This is a very, very bad idea. Please don’t do it.
Great thanks! I know he has been overwhelmed by primer questions before, but this is an open forum so future wonderers can just come here and look for answers.
Thanks for the tip Dennis. I am planning to start with 3 times the recommended amount. But I will think this over one more time before I plunge. Have kept this month to really think over the plans before diving in.
I don’t understand. I can see absolutely no good reason to split orders into two accounts, but plenty of reasons not to. You are basically throwing away all your weapons and then going to war.
Hmmm… You are making me think again. But here is what I have thought of:
Lets say you have 15 buys and one sell. You will have to pay for 13 buys when trading on a broker with zero margin for hedged trades. This can be countered if you choose a broker who offers high leverage. For eg with a 1:100 leverage broker if you give $100, then with a 1:500 broker you pay only $20. I am thinking that used margin is worst when you have a lot of buys and small sells or vice versa but by choosing a broker with high leverage I maybe can contain it with 3 times the starting balance.
The swap savings as I am choosing will give me a 1:3 advantage, that means if i gain 3$ for buys i only pay 1$ for sells. this i think will add gains to the equity.
Let me know your opinions. I guess i also have to think of the future when I will have to increase the trade sizes.
CC,
was there an easy fix to the 1.4 to get it to stay on a fixed lot size,
because im thinking of letting that version run a live channel for me (with auto primers turned off), and then doing manual primers… (atleast until vi have some fully working primers).
and you made it so in the newer versions, that the EA would look at old trades, and make extra to get the right combined lot amount… has this disapered again in the 1.4 version?
You’re misunderstanding things. if you have 15 long, 0 short, you pay for 15 long. With 15 long, 10 short you pay for 5 long in the same account, 25 (15+10) in two accounts. That’s the margin weapon of hedging.
Leverage would make no difference between 1 and 2 accounts.
Swap will make no difference between 1 and 2 accounts.
I take that back. You will both pay swap (difference in interest rate between two currencies) and earn from swap. If all earnings go into one account and all payments are deducted from the other, there will a difference, but you won’t like the outcome.
