Expanding Grid | Page 28 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Grid - New indicator that replaces chart grid 60 replies

Grid (not grid system) or horizontal lines indicator 6 replies

Scalping Expanding Hedged Grid 70 replies

Grid, Burst Grid, Grid Risk and Grid Profit 1 reply

Double Time Grid and Pip Grid 0 replies

Attachments: Expanding Grid
Exit Attachments

Expanding Grid

  • #541
  • Dec 15, 2015 10:31pm Dec 15, 2015 10:31pm
yes i am trying to work on the "hedge" ratio too , but a good mathematician or options trader with volatility computation will be able to get it faster

Quoting bauta
{quote} Yes, but they will open and close trades at its own set, you may have 10 lots in one set and only 0.01 in the other. style Multi_2 is symmetry and Floppy, they do trade individually too, the difference is that if 120% profit is reached, it will close both sets. I will be testing a counter idea where the initial quantity set automatically to half of the loosing set. example, if symmetry accumulated 0.8 lots long, mChannel then opens 0.4 lots short. In that scenario the account balance a bit. (work in progress idea) I do not opens the full...
  • #542
  • Dec 15, 2015 11:45pm Dec 15, 2015 11:45pm
good results settings:
going back to sExpandGrid, I did test lot=0.1, GridFactor=10 and moneyTakeProfit=1
Results are not bad at all, I will demo some of my explorers to see how it end up.
  • #543
  • Dec 15, 2015 11:57pm Dec 15, 2015 11:57pm
I support the idea of NOT changing TP as the basket grows.

The reason: when we have 5+ grid levels in trade, it means that we are already fighting a strong trend for sure (like today).

In other words: Why to risk the possibility of getting stuck in a longer countertrend? Just to squeeze out a few more bucks from the basket?

It is simply not worth the risk of ending up with an exponentially larger basket AND missing out on fresh and safer trades.

It is better to get rid of all large baskets as soon as possible (with standard profit), to take a deep breath of relief, then RINSE and REPEAT...
  • #544
  • Dec 16, 2015 12:27am Dec 16, 2015 12:27am
Hi Bauta , I'm running ea on live demo acounts this week as symmetry, symmetry2 and scalper trade types .results to now are in additional file.
Attached Image (click to enlarge)
Click to Enlarge Name: Adsız1.png Size: 117 KB
https://tracking.avapartner.com/_RwAAA
  • #545
  • Dec 16, 2015 12:49am Dec 16, 2015 12:49am
Running LIVE demos have the only advantage of checking EA functionality and getting new ideas.

On the other hand, to see the frequency and impact of WORST CASE SCENARIOS, high quality backtests are needed, which have been done by Master Bauta.

We only need to take care of the HIGH DRAWDOWN scenarios (e.g. by limiting trading hours), and the profits will take care of themselves.
  • #546
  • Dec 16, 2015 12:53am Dec 16, 2015 12:53am
Ok Guru, I staying tunned to thread...
https://tracking.avapartner.com/_RwAAA
  • #547
  • Dec 16, 2015 1:09am Dec 16, 2015 1:09am
test for 'black swan' type movements trying on GN
  • #548
  • Dec 16, 2015 1:44am Dec 16, 2015 1:44am
Quoting rabbit68
test for 'black swan' type movements trying on GN

That is exactly what I meant above.

We should concentrate on the worst case scenarios IN ADVANCE, because this strategy's Stop Loss is the MARGIN CALL.
  • #549
  • Dec 16, 2015 2:31am Dec 16, 2015 2:31am
Quoting MasterGuru
{quote} That is exactly what I meant above. We should concentrate on the worst case scenarios IN ADVANCE, because this strategy's Stop Loss is the MARGIN CALL.
You are absolutely right in general. However, after spending a lot of time on comparable grid recovery strategies i have to tell you that a timeframe limitation won't save you from strong trendings. Some heavy movements without retracement started during asian session even in sleeping pairs. I can look for examples if u want me to. It reduces the risk of course, but leaves out gains as well.

I think the focus should be on getting out of the basket as soon as possible. In this regard it may be an option to have an exponential factor (instead of doubling) to pull the breakeven point closer once a certain basket level is reached. To compensate the lotsize increase a "partly" hedge order could be set in parallel like the famous swb_grid did. This partly hedge should then be treated (recovered) seperately once the heavy grid reached breakeven.
Attached File(s)
File Type: ex4 swb grid 4.1.0.3.3.1_h_plus.ex4   36 KB | 513 downloads
  • #550
  • Dec 16, 2015 5:41am Dec 16, 2015 5:41am
Some really good posts and ideas in this thread now.

I spent years using Blessing 3 and super money grid (which does have a hedge/balance parameter) and have to agree with guru and drolph regarding closing a basket as soon as possible the higher the level open.
When it gets to level 5 and above closing the basket for a minimal profit is preferable in my opinion, just cover the commission/swap/spread.
Small regular profits is the way to go.
  • #551
  • Dec 16, 2015 6:26am Dec 16, 2015 6:26am
Quoting Drolph
{quote} You are absolutely right in general. However, after spending a lot of time on comparable grid recovery strategies i have to tell you that a timeframe limitation won't save you from strong trendings. Some heavy movements without retracement started during asian session even in sleeping pairs. I can look for examples if u want me to. It reduces the risk of course, but leaves out gains as well. I think the focus should be on getting out of the basket as soon as possible. In this regard it may be an option to have an exponential factor (instead...

@Drolph

I do accept that some major moves could happen during "sleeping hours", as e.g. natural disasters and disasters in general don't care about market times... Nevertheless, it would be interesting to see one. If you have the time please show us an example.

What Master Bauta does in his forward tests - i.e. closing all trades before major market news - should have significant positive impact on minimizing DD, however it would be practically impossible to model by backtesting. Matching the starting points of major DDs in the backtests with past Red News Events could be a reassurance; but again, how do we dig up Red News Events going back years...?

Finally, using an exponential factor for "runners" might be very helpful, which I would use relatively early, as beyond the 5th grid, the impact of an expontential factor would very signficant. As @Nomrtold also mentioned, the goal would be getting out of baskets ASAP at level 5 or above, as the basket is getting into dangerous territories. From this aspect the "Scalper" style may be the best approach and setting for this EA.

Personally I am not a fan of hedging, mainly because we are substituting one problem with another, so I would concentrate on making the strategy "Black Swan Proof" and "Red News Proof".
  • #552
  • Dec 16, 2015 7:12am Dec 16, 2015 7:12am
It might not be the best example but the first one i found by quick search.

EURGBP runner from 2010 started during "sleeping time" on Feb 25. In the run shown it recovered luckily just before weekly close. However by changing the starting day some grids were badly placed, so it suffered the additional 130 pips move at the weekly open leading to a DD of over 10k.

Although there were some labor data during these days, there were no high impact ECB session or anything comparable.
Attached Image (click to enlarge)
Click to Enlarge Name: example.png Size: 17 KB
  • #553
  • Dec 16, 2015 7:53am Dec 16, 2015 7:53am
Quoting Drolph
It might not be the best example but the first one i found by quick search. EURGBP runner from 2010 started during "sleeping time" on Feb 25. In the run shown it recovered luckily just before weekly close. However by changing the starting day some grids were badly placed, so it suffered the additional 130 pips move at the weekly open leading to a DD of over 10k. Although there were some labor data during these days, there were no high impact ECB session or anything comparable. {image}
Thanks for the example. With a magnifier it seems that the run-up has started around 9 AM (on Feb 24th) and - by using Scalping Style - all trades could have been closed with profit during the early hours of the Asian Session, and definitely by 6:30 AM. The same would apply for the Asian Session of February 25th.

In my previous proposal I suggested ONE HIGH STAKE trade during the Asian session limiting fund exposure to a very narrow time frame. (Of course, it would need some confirmation by backtests).
  • #554
  • Dec 16, 2015 8:44am Dec 16, 2015 8:44am
Quoting normtold
Some really good posts and ideas in this thread now. I spent years using Blessing 3 and super money grid (which does have a hedge/balance parameter) and have to agree with guru and drolph regarding closing a basket as soon as possible the higher the level open. When it gets to level 5 and above closing the basket for a minimal profit is preferable in my opinion, just cover the commission/swap/spread. Small regular profits is the way to go.
interesting idea,
I did add a parameter as BreakevenLevel=5 per example, then EG will set target to 0.01 if grid levels are above it and till cycle closes.
Let's see
Attached File(s)
File Type: ex4 EG103.ex4   76 KB | 325 downloads
  • #555
  • Dec 16, 2015 8:46am Dec 16, 2015 8:46am
ALERT:
I do setting ALL pairs to CLOSURE MODE starting right now.
Set Target to 0.1 and CONTINUE=false;
resume two hours after FOMC release.
bauta
  • #556
  • Edited 12:08pm Dec 16, 2015 10:32am | Edited 12:08pm
Quoting bauta
{quote} interesting idea, I did add a parameter as BreakevenLevel=5 per example, then EG will set target to 0.01 if grid levels are above it and till cycle closes. Let's see {file}
BreakevenLevel=5 test results.
This parameter helped in the drawdown by more than 40%, also profit factor increased.
Note that in this revision I did removed symmetry2, if you like to trade symmetry2, just use symmetry with Grid_Increments=30 and MoneyTakeProfit=2000

Now, capital needed is 25K instead of 50K

Please re-download at post 1
http://www.forexfactory.com/attachme...1&d=1450283497

Test results:
Attached Image (click to enlarge)
Click to Enlarge Name: Image1.png Size: 39 KB
  • #557
  • Dec 16, 2015 10:44am Dec 16, 2015 10:44am
"Now, capital needed is 25K instead of 50K"

That is great news.
Will it work on a live account?
I would feel a bit calmer using it than the version I am using live now.
  • #558
  • Dec 16, 2015 10:47am Dec 16, 2015 10:47am
thank you sir, will retest with this version once the attachment uploads and appears on page 1....

BreakevenLevel=5 test results.
This parameter helped in the drawdown by more than 40%, also profit factor increased.
Note that in this revision I did removed symmetry2, if you like to trade symmetry2, just use symmetry with Grid_Factor=1 and MoneyTakeProfit=2000

Now, capital needed is 25K instead of 50K

Please re-download at post 1
http://www.forexfactory.com/attachme...1&d=1450283497
  • #559
  • Dec 16, 2015 11:18am Dec 16, 2015 11:18am
@Bauta

Interesting that the DD has improved so much just by a small adjustment of the MoneyTakeProfit parameter.

What if we accepted a small loss from the 6th level by setting MoneyTakeProfit at -1.

It would be a small price to pay for getting out of potential large drawdowns (before they happened), which DDs mean not only anxiety, but also LOTS OF MISSED PROFITABLE OPPORTUNITIES.

In other words: getting out of a "bad basket" IN TIME - with a small loss - could be even profitable by allowing the EA to mak numerous profitable trades which otherwise would have been missed, i.e. while waiting for the "bad basket" to close with profit (above 5th level).

I wonder if you could run a set of backtests with negative (e.g. -1) MoneyTakeProfit and BreakevenLevel set at 5 or 6. Thanks!
  • #560
  • Dec 16, 2015 11:46am Dec 16, 2015 11:46am
I look forward to bautas view but in my opinion it will not make a lot of difference.

Level 5 = at least 0.16 Lots and the difference between + 1 cent and - 1 cent is a fraction of a Pip.
Even closing the basket at + 1 cent might still end up a small loss due to fast moving price or slow execution.
2 traders viewing now