The Ghillyoar System (buying highs and selling lows) | Page 8 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Daily Highs and Lows to Previous Daily Highs and Lows 1 reply

Higher Highs/ Lows and Lower Highs/ Lows 14 replies

buying and selling the news, eurusd 29/4/08 1 reply

Difference between selling a Call and buying a Put? 5 replies

Buying Dips and Selling Rallies 16 replies

Attachments: The Ghillyoar System (buying highs and selling lows)
Exit Attachments

The Ghillyoar System (buying highs and selling lows)

  • #141
  • May 3, 2021 2:41pm May 3, 2021 2:41pm
Quoting neosqualls
Hi JoanGuz, You have done great work I totally agree about the swap and unrealized drawdown. Do you think the bench should be adapted to average range ? Also here is an idea coming from a friend of mine : Exit rule #3 : Exit if account equity is less than highest equity * 95%. Reset the highest equity level after the exit. Regards.
Hi Neosqualls!

Thanks for the feedback and ideas!!

I do think that the bench should be dynamic, and the average range is a great place to start. I have manually changed the default (30) value, sometimes works great with a smaller value and sometimes with a bigger one. So, adding the range should be better to adapt to differet currencies with distinct volatility.

About Exit Rule #3, I added that exact rule in the weekend and the results were worse almost everytime, I say "almost" but I don't remember any pair that improved the results jeje. Look an example, first image is the system with default rules, and the second one is with Rule#3, parameters: EURSUD 2017-2021, Daily TF. Its worse because the rule realizes losses that will turn profitable ahead.

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSD Daily Bench.gif Size: 9 KB

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSD Daily With MaxLoss.gif Size: 10 KB

Right now I'm experimenting the system without rule #2, and looks super interesting! Obviously with smaller TimeFrames (to let the recovery happen faster) and smaller Lot Sizes, because the system sometimes opens a lot of trades before hitting the target. I'm doing that because I do think that we need a rule to exit when the system is flat and opening a lot of trades without going anywhere, but this rule sometimes closes losers with a terrible timing and realizing big drawdowns.

Regards!
  • #142
  • May 4, 2021 11:53am May 4, 2021 11:53am
Quoting JoanGuz
{quote} Hi Neosqualls! Thanks for the feedback and ideas!! I do think that the bench should be dynamic, and the average range is a great place to start. I have manually changed the default (30) value, sometimes works great with a smaller value and sometimes with a bigger one. So, adding the range should be better to adapt to differet currencies with distinct volatility. About Exit Rule #3, I added that exact rule in the weekend and the results were worse almost everytime, I say "almost" but I don't remember any pair that improved the results jeje....

removing rule#2 you let the system only exit when in gain, so you implement a plane antigrid system...which suffers of long/indefinite "frozen" periods with the cross wandering between the average long price and the average short price.


dynamically redefining the benches can unblock the "system freezing"...but what rule have you used to recompute the benches?
  • #143
  • Edited 11:28am May 5, 2021 11:00am | Edited 11:28am
Quoting tesiag
{quote} removing rule#2 you let the system only exit when in gain, so you implement a plane antigrid system...which suffers of long/indefinite "frozen" periods with the cross wandering between the average long price and the average short price. dynamically redefining the benches can unblock the "system freezing"...but what rule have you used to recompute the benches?
I totally agree, I was testing how the system does without rule#2, to watch under what circumstances it tends to get "frozen" and think a for a better alternative, which I haven't found.

The rule that I have used to recompute the benches is to record when the "Last Bench" (could be the Sell or Buy, either count) was recalculated, and if X amount of bars pass, I used the original rule to add / substract the pips from Candle 2.
  • #144
  • May 5, 2021 2:04pm May 5, 2021 2:04pm
Quoting JoanGuz
{quote} I totally agree, I was testing how the system does without rule#2, to watch under what circumstances it tends to get "frozen" and think a for a better alternative, which I haven't found. The rule that I have used to recompute the benches is to record when the "Last Bench" (could be the Sell or Buy, either count) was recalculated, and if X amount of bars pass, I used the original rule to add / substract the pips from Candle 2.
Hi Joan,

So far from my tests, the risk/reward ratio with reasonable drawdown is not good. 90% of the time profit will be touch fast and secure but the 10% remaining will kill the concept.

As i already said backtest doesn't proof a strategy will work but proof a strategy would have worked in the past...

Having multiple position open at the same time for a long term will cost more than it pays.

So what about killing an order after X period for example ?
  • #145
  • May 5, 2021 3:15pm May 5, 2021 3:15pm
Quoting neosqualls
{quote} Hi Joan, So far from my tests, the risk/reward ratio with reasonable drawdown is not good. 90% of the time profit will be touch fast and secure but the 10% remaining will kill the concept. As i already said backtest doesn't proof a strategy will work but proof a strategy would have worked in the past... Having multiple position open at the same time for a long term will cost more than it pays. So what about killing an order after X period for example ?
....or implement a "classical" grid (not antigrid) sized so as to:
1- compensate with some gains the "dead" times (and swap expenses) when the original system is "frozen" and
2- not to loose too much when the cross decides to lift off again
  • #146
  • May 5, 2021 3:36pm May 5, 2021 3:36pm
Quoting tesiag
{quote} ....or implement a "classical" grid (not antigrid) sized so as to: 1- compensate with some gains the "dead" times (and swap expenses) when the original system is "frozen" and 2- not to loose too much when the cross decides to lift off again
You want to remove the interesting part of this. The acceptancy of loss is needed, no free lunch theoreme.

From what i found so far the profit factor is very low. Check paddy's backtest eurusd 1.12.19 to 10.3.21 arround 3000$ with 100k capital for this period.... Risk is high return is low. What do you guys think?
  • #147
  • Edited 4:33pm May 5, 2021 4:10pm | Edited 4:33pm
the return is very low......why??? like I said in post 126, you`re all concentrating on a few pairs, whether they are trending or not......so the profits are good when its trending, losses when price is not going anywhere....what is so difficult to understand???
so what should you all be doing?!?.....concentrating on instruments which are trending....
you might be trading 5 trending instruments and switch to others when 1 is stalling...
you`re making it far too difficult....no need for an EA....keep an eye out for trends on a weekly basis.... look at your trades once a day with the occasional trade thrown in during the week.....
you dont have to do much......the trend will do it for you.....
this way is only for lazy people, who are not trigger happy....but have enough energy to look for trends on a weekly basis...
but not lazier than lazy and only want an EA to trade for them...
that`s all it takes to make major profits.....
  • #148
  • May 5, 2021 6:13pm May 5, 2021 6:13pm
Quoting yonnie
the return is very low......why??? like I said in post 126, you`re all concentrating on a few pairs, whether they are trending or not......so the profits are good when its trending, losses when price is not going anywhere....what is so difficult to understand??? so what should you all be doing?!?.....concentrating on instruments which are trending.... you might be trading 5 trending instruments and switch to others when 1 is stalling... you`re making it far too difficult....no need for an EA....keep an eye out for trends on a weekly basis.... look...
You are certainly right in principle...but what about your exit (stop loss) management? When do you think the trend is not in place anymore? Or how large needs to be' the loss acceptance?
  • #149
  • May 5, 2021 8:03pm May 5, 2021 8:03pm
I`m not going into how to trade this much.....most of you are just readers and not traders....certainly not profitable traders....so I wont waste my time...
but for the few who really want to learn, go to "Grid trading system, keeping DD to a minimum" and you`ll see a few examples....to make it even easier, you can trade only long in a uptrend and short in a downtrend....
but yeah, you got to put some effort in to start and that is already too much for most people......
  • #150
  • May 6, 2021 2:25pm May 6, 2021 2:25pm
Quoting yonnie
the return is very low......why??? like I said in post 126, you`re all concentrating on a few pairs, whether they are trending or not......so the profits are good when its trending, losses when price is not going anywhere....what is so difficult to understand??? so what should you all be doing?!?.....concentrating on instruments which are trending.... you might be trading 5 trending instruments and switch to others when 1 is stalling... you`re making it far too difficult....no...
Quote
I`m not going into how to trade this much.....most of you are just readers and not traders....certainly not profitable traders....so I wont waste my time...
but for the few who really want to learn, go to "Grid trading system, keeping DD to a minimum" and you`ll see a few examples....to make it even easier, you can trade only long in a uptrend and short in a downtrend....
but yeah, you got to put some effort in to start and that is already too much for most people......

Why is the return very low ? if it's not clear enough ... i can't do anything more
Who said i am concentrating on few pairs?
Are you talking about you?

Best regards.
  • #151
  • May 10, 2021 3:07pm May 10, 2021 3:07pm
Attached Image (click to enlarge)
Click to Enlarge Name: Ghillyoar tweak 10 5 21.gif Size: 13 KB
  • #152
  • May 11, 2021 4:47am May 11, 2021 4:47am
Attached Image (click to enlarge)
Click to Enlarge Name: Another tweak.gif Size: 13 KB
  • #153
  • May 11, 2021 10:09am May 11, 2021 10:09am
Quoting Paddy9
{image}
what do those backtest refer to?
I mean start and end time, EA parameters etc


thanks
  • #154
  • May 11, 2021 11:03am May 11, 2021 11:03am
Quoting tesiag
{quote} what do those backtest refer to? I mean start and end time, EA parameters etc thanks

My apologies, I should have said. Both are:

eurusd: 1.1.2000 to 23.3.2021.
Ghillyoar : with a tweak.
  • #155
  • May 11, 2021 11:04am May 11, 2021 11:04am
Exit rule number two states:

Exit Rule Number Two
One of the two following conditions has to be satisfied.

The total Open profit on buys has to be higher than 20$.
or
The total Open profit on sells has to be higher than 20$


In addition the following condition must be satisfied:
The number of open buys must be equal to the number of open sells.

Trying to understand why is this condition "The number of open buys must be equal to the number of open sells" paramount! For example, in the image attached below (demo of the EA), GU has a sell that is -£121.23, and 3 buys that are in total +£132.68, so +£11.45 profit overall. So, would it not make more sense to close at an overall total profit (off both sells and buys irrespective of the number of buys or sells) instead of waiting until number of buys equals number of sells? Just wondering why this condition by Paddy9. I am obviously missing something!

Attached Image (click to enlarge)
Click to Enlarge Name: example.JPG Size: 225 KB
  • #156
  • May 11, 2021 11:06am May 11, 2021 11:06am
Quoting Paddy9
{quote} My apologies, I should have said. Both are: eurusd: 1.1.2000 to 23.3.2021. Ghillyoar : with a tweak.
What is the tweak? Thanks
  • #157
  • May 11, 2021 11:52am May 11, 2021 11:52am
Quoting danjuma
{quote} What is the tweak? Thanks
maybe the use of a higher lotsize (not 0.1 but 1)


....mhhhh...no....there's also something which makes drastically increase the frequency of the trades (around 1200 trades in 1 year)
  • #158
  • May 11, 2021 11:58am May 11, 2021 11:58am
Quoting tesiag
{quote} maybe the use of a higher lotsize (not 0.1 but 1) ....mhhhh...no....there's also something which makes drastically increase the frequency of the trades (around 1200 trades in 1 year)
It is 1200 trades in 20 years my friend.

Probably the TargetStep is not fixed but based on account balance as percentage
  • #159
  • May 11, 2021 12:13pm May 11, 2021 12:13pm
Quoting neosqualls
{quote} It is 1200 trades in 20 years my friend. Probably the TargetStep is not fixed but based on account balance as percentage
you're right ...sorry
  • #160
  • May 11, 2021 12:22pm May 11, 2021 12:22pm
Quoting neosqualls
{quote} It is 1200 trades in 20 years my friend. Probably the TargetStep is not fixed but based on account balance as percentage

don't think so...in such case I would expect an exponential (at least a non linear) growth of the equity...like when you apply money management techniques to a succesful strategy

the equity growth is completely linear...so fixed parameters..ora at least not linked to something constantly growing
0 traders viewing now