Donchian Channel Trading

[kingalpha]

Bob, have you recent made changes your live test account? There seems to be a lot of trades lately but average trade duration is much shorter than a few weeks ago.

[_bob]

Bob, have you recent made changes your live test account? There seems to be a lot of trades lately but average trade duration is much shorter than a few weeks ago.

Well spotted bro.

Yep things have changed I have changed. I am retiring from Donchian Channels. I’ve had great fun developing a few systems and running them through the processes. The work that is put in by others and the knowledge they posses I admire greatly. Anybody starting their journey wouldn’t go wrong starting here.

But as a trader I have to be true to myself. And I love my tick charts. My method is very simple, very disciplined and very rewarding. It is also extremely boring, mind-blowing repetitive and wide open for improvement. My brain freeze a couple of weeks ago was a wake up call and its time to go back and redesign the system. Unfortunately I have limited capital, you can’t back test on tick charts and demo trading doesn’t cut the mustard down at this level. So a small live account, one that you are prepared to destroy is a prerequisite at this level. You simply have no choice.

So do I use this account to continue forward testing DC. I have no doubt that this is the more guaranteed method. Or do I take the risk, use the knowledge I posses on ticks, combine it with the skills I have learnt recently and head back. Got to be true to myself bro. And if you take no risks, you get no rewards.

So to me it’s a no brain-er. The best thing about being an armchair trader is that there are simply no rules. You answer to no-one. And the only thing you thing you kid is your account balance. And as it is above so too below.

Happy trading all

Bob

[rihan]

Forget about pips and think in terms of percentages. Position sizing is the most overlooked part of trading.

Two Donchian Channel traders buy EUR/USD at 1.1500 and place stops at 1.0000. They both have 1500 pips in risk from entry to stop. Trader Joe put on one micro (1,000) so his dollar risk is $150. His account size is $1500 so he is risking 10% of his account on the trade.

Trader Jane has an account size of $1500 also but she just wanted to risk 1% so she just put on 100 units. Two weeks later EUR/USD is at 1.0500 and both traders are down 1000 pips in the trade but Trader Joe is down $100 which is 6.67% of his account while Trader Jane is down just $10 which is 0.67% of her account.

In addition to that, Trader Jane put on 9 other trades in different markets that each risked 1% so her total open risk was the same size as Trader Joe’s but her likelihood of being in at least on winning trade was 10 times more than Trader Joe’s.

Trader Jane is a systematic diversified trend following trader. Trader Joe is uses Donchian Channels like Trader Jane but does not diversify his risk in different markets and does not concern himself with risk parity (position sizing). That is the difference.

-Adrian

I am currently reading Michael Covel’s Trend Trading. I have changed my perception to rather see a % increase/decrease than pips. One thing I am battling with : It is said that dd is normal, investors liked it because they could get on for much less, but how do we differentiate between normal drawdown and a losing trade? What makes it so successful is to CUT YOUR LOSERS SHORT. Is it determined as a losing trade when it hits our SL?

[Arbitrager_on_Acid]

I am currently reading Michael Covel’s Trend Trading. I have changed my perception to rather see a % increase/decrease than pips. One thing I am battling with : It is said that dd is normal, investors liked it because they could get on for much less, but how do we differentiate between normal drawdown and a losing trade? What makes it so successful is to CUT YOUR LOSERS SHORT. Is it determined as a losing trade when it hits our SL?

The initial stop loss is set at the point of entry into the trade. In a pure Donchian Channel system, that stop is simply set at the first pip or tick outside of the channel in the direction against the trade. So if You go long at the break of a channel to the upside, you have a stop set just below the channel to the downside. Some traders (the Turtles included) use an initial stop distance that is based on something else like ATR (that is what the Turtles used) but of course a pure Donchian Channel system would just use the opposite side of the channel for the initial stop placement.

The more important consideration is next: Once you have determined that you will get in at a certain point and that you will place your initial stop at a certain point, you must set your position size. You may know for example that you are going to buy EUR/USD at 1.1500 with a stop at 1.1000. But how many units will you buy? 1,000? 10,000? 3,217? 2?

Most trend following traders (the Turtles included) choose a position size that will set the percentage of their equity which they will lose if their initial stop is taken out to a certain consistent amount such as 0.33% (33 basis points). (One basis point is one one-hundredth of a percent. Commonly, a number of basis points such as twelve basis points is written out as: “12bps” and pronounced “twelve bips”).

Examples:

0.25% is zero point two five percent. It is 25 basis points and can be written “25bps”.
0.12% is zero point one two percent. It is 12 basis points and can be written “12bps”.
0.59% is zero point five nine percent. It is 59 basis points and can be written “59bps”.
0.31% is zero point three one percent. It is 31 basis points and can be written “31bps”.

A trend following trader may decide to risk 25bps per trade. So in the example above, the trader will set a stop entry at 1.1500 with a stop loss at 1.1000 and if his account is worth $10,000 he will want a position size that will risk $25 from entry to stop loss (10,000 x 0.0025 = 25, 25 is 25bps of 10,000). Because this trade puts 500 pips between the entry and the initial stop loss (11500-11000=500) he will want to risk $0.05 per pip (25/500=0.05). This particular pair will risk one cent per pip per 100 units traded. So if he puts on 500 units of EUR/USD with that entry and that stop, he will risk $0.05 per pip over 500 pips for a total risk of $25. So his position size will be 500 units.

If this trade is stopped out at the initial level, he will lose $25. He may have 40 trades all risking that same $25 amount going at once, but all of them risk that same $25 no matter what the pip distance is between his entry and his initial stop because he sets the position size to make the risk from entry to stop that same $25 amount. This gives him what many call “risk parity” which means he risks the same amount on every trade despite the pip distance between entry and stop.

So if this trader has ten straight losses in a row, he will lose 25bps or less per trade (plus any slippage) and thus accumulate less than a 2.5% drawdown as a result of those ten losses. Many times the stop will be groomed to a position wherein the trader will lose less than the initial 25bps, but it will not be groomed to a profitable position. Under such circumstances he will have a losing trade, but one that loses less than the 25bps. In that case he has a drawdown, but one smaller than his initial risk per trade.

The stop loss is never removed and allowed to stay in the market until it is taken out. It is simply groomed along the bottom of the channel on a long trade or the top of the channel on a short trade until it is taken out. This keeps your losses smaller than your wins. Your losses are limited to known small amounts while your gains can go on and on to win many times your initial risk. So one winning trade can win more than the accumulated losses from ten losing trades. That is the edge this type of system has: asymmetrical reward to risk ratios.

All drawdowns would be considered “normal” unless the trader overrode his system and put on a position size that would risk more bps than his predetermined level or he moved his initial stop against his trade.

When you look at the equity curve of a trader doing this kind of trading, you will see a smooth downward line as he loses many trades that give him very similar sized losses. But then that smooth line down will be interrupted by a big jump upward when big winners are closed. Then you will see another smooth decline followed by another big jump up.

-Adrian

[kingalpha]

When you look at the equity curve of a trader doing this kind of trading, you will see a smooth downward line as he loses many trades that give him very similar sized losses. But then that smooth line down will be interrupted by a big jump upward when big winners are closed. Then you will see another smooth decline followed by another big jump up.

-Adrian

Perfect description of what my account is actually doing

[rihan]

The initial stop loss is set at the point of entry into the trade. In a pure Donchian Channel system, that stop is simply set at the first pip or tick outside of the channel in the direction against the trade. So if You go long at the break of a channel to the upside, you have a stop set just below the channel to the downside. Some traders (the Turtles included) use an initial stop distance that is based on something else like ATR (that is what the Turtles used) but of course a pure Donchian Channel system would just use the opposite side of the channel for the initial stop placement.

The more important consideration is next: Once you have determined that you will get in at a certain point and that you will place your initial stop at a certain point, you must set your position size. You may know for example that you are going to buy EUR/USD at 1.1500 with a stop at 1.1000. But how many units will you buy? 1,000? 10,000? 3,217? 2?

Most trend following traders (the Turtles included) choose a position size that will set the percentage of their equity which they will lose if their initial stop is taken out to a certain consistent amount such as 0.33% (33 basis points). (One basis point is one one-hundredth of a percent. Commonly, a number of basis points such as twelve basis points is written out as: “12bps” and pronounced “twelve bips”).

Examples:

0.25% is zero point two five percent. It is 25 basis points and can be written “25bps”.
0.12% is zero point one two percent. It is 12 basis points and can be written “12bps”.
0.59% is zero point five nine percent. It is 59 basis points and can be written “59bps”.
0.31% is zero point three one percent. It is 31 basis points and can be written “31bps”.

A trend following trader may decide to risk 25bps per trade. So in the example above, the trader will set a stop entry at 1.1500 with a stop loss at 1.1000 and if his account is worth $10,000 he will want a position size that will risk $25 from entry to stop loss (10,000 x 0.0025 = 25, 25 is 25bps of 10,000). Because this trade puts 500 pips between the entry and the initial stop loss (11500-11000=500) he will want to risk $0.05 per pip (25/500=0.05). This particular pair will risk one cent per pip per 100 units traded. So if he puts on 500 units of EUR/USD with that entry and that stop, he will risk $0.05 per pip over 500 pips for a total risk of $25. So his position size will be 500 units.

If this trade is stopped out at the initial level, he will lose $25. He may have 40 trades all risking that same $25 amount going at once, but all of them risk that same $25 no matter what the pip distance is between his entry and his initial stop because he sets the position size to make the risk from entry to stop that same $25 amount. This gives him what many call “risk parity” which means he risks the same amount on every trade despite the pip distance between entry and stop.

So if this trader has ten straight losses in a row, he will lose 25bps or less per trade (plus any slippage) and thus accumulate less than a 2.5% drawdown as a result of those ten losses. Many times the stop will be groomed to a position wherein the trader will lose less than the initial 25bps, but it will not be groomed to a profitable position. Under such circumstances he will have a losing trade, but one that loses less than the 25bps. In that case he has a drawdown, but one smaller than his initial risk per trade.

The stop loss is never removed and allowed to stay in the market until it is taken out. It is simply groomed along the bottom of the channel on a long trade or the top of the channel on a short trade until it is taken out. This keeps your losses smaller than your wins. Your losses are limited to known small amounts while your gains can go on and on to win many times your initial risk. So one winning trade can win more than the accumulated losses from ten losing trades. That is the edge this type of system has: asymmetrical reward to risk ratios.

All drawdowns would be considered “normal” unless the trader overrode his system and put on a position size that would risk more bps than his predetermined level or he moved his initial stop against his trade.

When you look at the equity curve of a trader doing this kind of trading, you will see a smooth downward line as he loses many trades that give him very similar sized losses. But then that smooth line down will be interrupted by a big jump upward when big winners are closed. Then you will see another smooth decline followed by another big jump up.

-Adrian

Thanks Adrian, this really sheds a lot of light. I also read somewhere that the original Donchian way all positions are closed after it reaches a 35 day age.
But I would like to follow suit and keep positions open until they are taken out by stops rather.

[rihan]

Adrian, if I look at your FXCM account which makes use of a 10 day channel, then it can be seen as an account with a steady decline the past 2 years. It also doesn’t make use of stop losses and has a unfavorable dd. Is it an experimental account?

Will the true Donchian trading I follow look like that account or the oanda one?

[kingalpha]

Adrian, if I look at your FXCM account which makes use of a 10 day channel, then it can be seen as an account with a steady decline the past 2 years. It also doesn’t make use of stop losses and has a unfavorable dd. Is it an experimental account?

There is always a stop loss when trading DC, otherwise it is impossible to calculate position size. If you look back a few pages, Adrian said that if all trades are closed immediately then the account would be in profits. There is a difference between account balance and account equity.

Equity = Balance + sum of all open positions (profits or loss)

Speaking to position size, I use account balance to work out my trading size. Has anyone experimented with account equity instead?

[Arbitrager_on_Acid]

Adrian, if I look at your FXCM account which makes use of a 10 day channel, then it can be seen as an account with a steady decline the past 2 years. It also doesn’t make use of stop losses and has a unfavorable dd. Is it an experimental account?

Will the true Donchian trading I follow look like that account or the oanda one?

That account was used for several systems at once and for experimenting with higher amounts of leverage until April 2015. Since then it has been just for a 10 day Donchian system. It does use stops but myFxbook seems to not list them. That may be because of some setting or something. I don’t know.

-Adrian

[rihan]

That account was used for several systems at once and for experimenting with higher amounts of leverage until April 2015. Since then it has been just for a 10 day Donchian system. It does use stops but myFxbook seems to not list them. That may be because of some setting or something. I don’t know.

-Adrian

Haha. Thank you Adrian, that paints a whole other picture! So glad I asked, I was super confused.

[3xfx]

Bob really but lots of time and effort into back testing, I’m surprised he is not trading the channels anymore.

[JamesBr]

Bob really but lots of time and effort into back testing, I’m surprised he is not trading the channels anymore.

Don’t know if he mastered it then…

[_bob]

Looks like we attracted a professional troll to the thread.

Don’t know if he mastered it then…

See james, thats where the problem lies. I havent mastered it. But unlike yourself (presumption here) I don’t have the two key ingredients for success. Time and money. These resources as sacred to me. So I have to be very specific where I use them. I’ll be back to master DC but for the moment I’ll be true to myself and go back to my style which I’ve deviated from.

In the mean time James, if you have nothing to contribute then best you look elsewhere to troll and leave the good people here alone.

And by judging your contributions here ar BP I would suggest you have nothing.

[tonyro44]

bobbillbrone, i think JamesBr likes to get into people skin but he means no harm with that. I bet he is a nice guy, it’s probably his way of having fun , right JamesBr ??

Looks like we attracted a professional troll to the thread.

See james, thats where the problem lies. I havent mastered it. But unlike yourself (presumption here) I don’t have the two key ingredients for success. Time and money. These resources as sacred to me. So I have to be very specific where I use them. I’ll be back to master DC but for the moment I’ll be true to myself and go back to my style which I’ve deviated from.

In the mean time James, if you have nothing to contribute then best you look elsewhere to troll and leave the good people here alone.

And by judging your contributions here ar BP I would suggest you have nothing.

[3xfx]

Bob, didn’t mean to have anyone come at u. Doch is not my main system either. Also I never heard bob say he was gonna master it I do remember him him saying that the thread took him back to the love of Doch trading. If anyone has but the time in reading this thread he has but a lot of time and effort into and has even shared many EA’s and asked for nothing in return.

[Forexmike]

Bob, didn’t mean to have anyone come at u. Doch is not my main system either. Also I never heard bob say he was gonna master it I do remember him him saying that the thread took him back to the love of Doch trading. If anyone has but the time in reading this thread he has but a lot of time and effort into and has even shared many EA’s and asked for nothing in return.

Totally agree…

Bob has been a, if not THE, main contributor to this thread…he has brought many new ideas and tested them…

I really hope that he will pass by the thread from time to time, although doesn’t trade DCs for now…

[_bob]

Lads, you don’t take another credit for yourselves.

All I wanted to bring was a bit statistical analysis which is limited to and reflective of my skills. And to demonstrate some of the work that goes into developing a strategy. The ideas all come from your inputs.

Its no-one fault but my own that my ego blow up. The market is the greatest reality check out there. We’ve attracted the attention of some members I have great respect for. So we know we’re doing something right. Those that know me know I love to hunt out trolls and marketers. It’s only natural that we would attract their attention to. So time to step back, go back to ones roots and start again.

I love ticks and want to and have been testing some ideas there. But lets not forget I still run a DC bot on the 4hr chart. JamesBr, you’re more than welcome to hunt that one out and critic it. So it won’t be long before I’ll be back. One thing we haven’t discussed is the use of fractals for stops and profit points. Particular on the 4 hr chart I have seen so strong arguments for their use.

But at the end of the day isn’t that all the likes of myself are doing here. Putting our hard work out there for criticing by those who have been blessed to have skills, knowledge and resources greater than our own. All in hope that we can improve our own skills and maybe help those around us.

Good times ahead lads

Bob

[tonyro44]

Well said bobbillbrowne ,at the end of the we all benefit from this thread, newbies and experienced a like. Thank you for your understand of some of us that have a short fuse .

Lads, you don’t take another credit for yourselves.

All I wanted to bring was a bit statistical analysis which is limited to and reflective of my skills. And to demonstrate some of the work that goes into developing a strategy. The ideas all come from your inputs.

Its no-one fault but my own that my ego blow up. The market is the greatest reality check out there. We’ve attracted the attention of some members I have great respect for. So we know we’re doing something right. Those that know me know I love to hunt out trolls and marketers. It’s only natural that we would attract their attention to. So time to step back, go back to ones roots and start again.

I love ticks and want to and have been testing some ideas there. But lets not forget I still run a DC bot on the 4hr chart. JamesBr, you’re more than welcome to hunt that one out and critic it. So it won’t be long before I’ll be back. One thing we haven’t discussed is the use of fractals for stops and profit points. Particular on the 4 hr chart I have seen so strong arguments for their use.

But at the end of the day isn’t that all the likes of myself are doing here. Putting our hard work out there for criticing by those who have been blessed to have skills, knowledge and resources greater than our own. All in hope that we can improve our own skills and maybe help those around us.

Good times ahead lads

Bob

[rihan]

Hi there.

I am at a tug of war about whether I should use ATR levels as stops, or place them at the opposite sides of the range. I have read good things about other traders that use ATR - it enforces the old saying of cutting losses short, but it can also get you out on a whipsaw which could have yielded excellent profits… What are your views on this?

[_bob]

G’day rihan

Comes down to your own needs bro. As you can see, we all do things slightly different. Time frames, entry channel, risk, stops, pairs, trade management. Difference in trading styles at the end of the day. We’re not institutes, we don’t have to “trade” like them. It’s one of the few advantages we have.

So IMO there’s a difference in speculators like myself and investors like learned master Adrian. Our philosophies need to be different. Pure speculation requires locking in profits and high win rates. Investors cut losses short and let profits run therefor can maintain a much lower win rate.

So how you determine your stops is up to you and where you feel you lie in that spectrum. Note in my testing I used ATR’s on the 1hr charts. Yet on the 4 Hr I run off an “inside channel”. I’ve also recently stated that I see a strong arguement to use fractals to determine last swing points for stops. This moves in-line with your suggestion of using the opposite sides of the range. Indeed if memory serves me correctly, Adrian who trades of weekly charts does use the range.

See a pattern forming. So it comes down to you. The only way to answer that is to test your therories. Number of ways, program a bot. Sit there paper/demo trade. Forward test.

Best of luck on your search. Hopefully a few of the other lads will share their experience.

Bob