The DIBS Method... No Free Lunch continues | Page 474 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

The DIBS Method - No Free Lunch Returns 13 replies

No Free Lunch but all the Free Coffee you can drink 736 replies

Peter Crown's DIBS method revisited 14 replies

DIBS Method EA 11 replies

Problem using Dibs EA 4 replies

Attachments: The DIBS Method... No Free Lunch continues
Exit Attachments

The DIBS Method... No Free Lunch continues

  • #9,461
  • Mar 26, 2010 3:15am Mar 26, 2010 3:15am
i took these three H1-IBs in the Yen.

the next IB is showing up now, but already 3 bars between.

i think the yen is the key currency for the next 12 month to play with for the long haul.

incredible week so far...
Attached Image (click to enlarge)
Click to Enlarge Name: IB-YEN.jpg Size: 206 KB
  • #9,462
  • Mar 26, 2010 6:11am Mar 26, 2010 6:11am
a nice IB in AUD/USD now.
Attached Image (click to enlarge)
Click to Enlarge Name: ib-aud.jpg Size: 261 KB
  • #9,463
  • Mar 26, 2010 6:32am Mar 26, 2010 6:32am
Quoting RussiaTrader
a nice IB in AUD/USD now.
I'm guessing that's a H1 countertrend? How far would you ride it?
  • #9,464
  • Mar 26, 2010 6:39am Mar 26, 2010 6:39am
Quoting MattW
I'm guessing that's a H1 countertrend? How far would you ride it?
well as the original rule, the first half @1:1 or better.
the second i would hold for months if i am not stopped out.
  • #9,465
  • Mar 26, 2010 6:53am Mar 26, 2010 6:53am
As you took that according to the rules...

I've not got a chart in front of me so apologies if I'm wrong - but if that IB is from 10am to 11am surely it's down on the day and you're looking for shorts, not vice versa?
  • #9,466
  • Mar 26, 2010 7:36am Mar 26, 2010 7:36am
yes exactly that, you are looking to buy strength and sell weakness.
  • #9,467
  • Mar 26, 2010 7:43am Mar 26, 2010 7:43am
Quote
1. how much data did you back test
2. Maybe you have a few too many rules. How do you know if any of these are valid or invalid rules?

1. I used tick data from 2001 with a 90% modelling quality.
2. I actually don't have too many rules.

Lets think about what the essence of the DIBS system is and then build a logical set of rules to define a system, again using ideas from Petercrown's posts.

"TRADE BREAKOUT OF A CONGESTION IN THE DIRECTION OF THE LARGER TERM TREND."

Quote
1. Only buy when inside bar closes above the weekly Open. Only sell when the inside bar closes below the weekly open.

Quote
1. OK, I can see where you are coming from. But I dont really know if it has an impact on your results. Especially when you have all the other factors

here is PC's posts on this idea.

A method is pretty hard to trust if it only works on one timeframe.

For the 4/6/8 hour charts and daily charts you check to see if you are up or down versus last week's close. Take buys if you are above, sells if you are below.

Quote
3. Only conside those inside bars that have a range of less than 70% of the prior bar.

Quote
3. Why?

This one was one of MY filters. The idea is the market should first go into congestion and therefore an IB and then we trade the Breakout of the IB. If the IB is more than 70% (or any high number) of the prior bar, it does not really signify congestion but more of an indecision. So I stay away from it.

Quote
4. if an inside bar straddles the weekly open, the bias will be to sell only when the inside bar low is below the 48 period EMA.

Quote
4. I can see the trouble with this situation. If the IB is below the moving average you are using for a trailing stop, you are obviously not going to go long. this is your buy/sell line

This is only for those IB's that have high above the weekly open and low below the weekly open. If I have an IB that is below the MA line but occurs in an "UP" week, I will take the IB long.

Quote
5. An inside bar once created will trigger an entry search. The inside bar will become invalid if a bar close above/below the hi/lo of the bar prior to the IB.

Quote
5. This could be your major issue. Let me see if this is what you mean. Say you get a short entry signal and then a new bar forms that has a higher high and higher low than your IB, you will not use the IB as an entry trigger at all? Or if a new bar breaks the high of the bar prior to the IB. Either way, this is probably killing a lot of entries for you

I picked up this rule from looking at IB's from PC's charts. Actually I am giving the IB lots of room to trigger. Say I get a short signal. I will now look to sell the break of the IB. lets call the bar prior to IB as Trend Bar. I will continue to look to sell the break of the IB till I get a close above the high of the Trend Bar. If a bar breaks the high of the trend bar but does not close above it, the IB does not get invalidated but will remain and active IB. Only a close above the Trend Bar invalidates the IB.

Quote
6. If price has travelled more than 30% weekly ATR on long side, short entries will not be taken in that week. Conversely, if the price has travelled more than 30% weekly ATR

Quote
6. Probably way too complicated - over kill. it looks like you are trying not to lose

The thought process behind this rule is that if the market is going to reverse after having moved considerably in one direction in any given week, it is better to just stand aside that week.

Quote
8. No trades during early Monday or Late Friday.

Quote
8. Why? Especially if you are playing longer term. The only IB i will not use is the one that forms sunday evening est

These are times of very low liquidity and hence best avoided. Much the same reason as PC only trades IB's occurring within the most active hours of the day.

Quote
9. a new trade will be placed only when the prior trade has reached at least 1:1 RR distance. The prior trade does not have to be risk free.

Quote
9. Why?

Risk management. If I have an open trade, I will not take another trade in the same market unless my open trade is risk free. The rule is identical to taking half of at 1:1 except that we are not closing any position but just stop looking for a new one till 1:1 happens.

Quote
10. Trades will be closed with a trailing stop. for a long trade, if the close is below the 48 EMA, the stop will be updated to the low of the bar that broke the MA line.
For a short trade, if the close is above the 48 EMA, stop will be updated to above the bar that broke the MA line.

Quote
10. your trailing stop is too complicated. you may be letting a trade run too much against you. why not just make it simple. Simply us a close below an MA.

This one does not impact the results.

I know PC and another member have done some modelling and claim that this method has a profit factor of around 1.2. I am just unable to get that kind of results consistently across different markets and market conditions.

refer to this post for the modelling done by PC..

http://www.forexfactory.com/showthre...67#post2616367
  • #9,468
  • Mar 26, 2010 7:53am Mar 26, 2010 7:53am
Quoting MattW
You're right, you cant automatically back-test a discretionary system. But you can manually - either by scrolling back the chart and "taking trades" as they come into view, or by buying some FX strategy testing software that allows you to trade on historical tick data.
If you can define the rules for a system, you can automate it and back test it. Manually back testing and automatically back testing have the same drawbacks though manual backtesting has an additional flaw on not being able to test multiple markets and market conditions over a very large period of time. If you are referring to eyeballing the trades by scrolling left on the chart, then I think it is a complete waste of time. It will never just give you what you want to see and not the reality.

Quote
How do you know that a discretionary system will keep working? Well, of course you don't, any more than an amazing back-tested automatic system will. But you'll be trading with small risk so if it looks like the method has stopped working you change the method, right?

I don't. But with a backtested system, I know when the system is going out of whack based on my test results. With an disretionary system, there is no way to consider that the system might not be working as desired except just using a very random number of a drawdown of say 25%.

Quote
But will inside bars really stop working? What do you think?

I have no doubt whatsoever that the idea of IB's with long tail trades will work. I just am not able to prove it to myself much like PC and others have done it for themselves.
  • #9,469
  • Mar 26, 2010 8:06am Mar 26, 2010 8:06am
Quoting Eklavya
If you can define the rules for a system, you can automate it and back test it.
That's my point. A discretionary system by definition does not have a complete set of rules. As an example, say your system dictated a TP at 1R, which happens to be 100 pips for a given trade. A defined ruleset would say TP at 100 and if price reached 98 pips and reversed, you would end up with a BE or a Loss depending on your stop rules. Whereas a discretionary system would allow the trader to see price reversing and close out for a partial profit at, for instance, 85 pips.

Quoting Eklavya
If you are referring to eyeballing the trades by scrolling left on the chart, then I think it is a complete waste of time. It will never just give you what you want to see and not the reality.
No, I meant scrolling right on the chart. So each single bar comes into view one at time. It is not perfect, of course, because they are closed bars - but it's better than nothing and should give you a reasonably accurate picture.

Quoting Eklavya
With an disretionary system, there is no way to consider that the system might not be working as desired except just using a very random number of a drawdown of say 25%.
Agreed.


  • #9,470
  • Mar 26, 2010 8:41am Mar 26, 2010 8:41am
Quoting MattW
As you took that according to the rules...

I've not got a chart in front of me so apologies if I'm wrong - but if that IB is from 10am to 11am surely it's down on the day and you're looking for shorts, not vice versa?
yes, but its a discretionary system and i stretch the rules due to my experience of trading with IBs.
the AUD trade had a pin bar first and then the IB.
in these cases i go long even when the day is negativ.
the trade is already closed break even.
  • #9,471
  • Mar 26, 2010 9:10am Mar 26, 2010 9:10am
[quote=MattW;3587451]
Quote
That's my point. A discretionary system by definition does not have a complete set of rules. As an example, say your system dictated a TP at 1R, which happens to be 100 pips for a given trade. A defined ruleset would say TP at 100 and if price reached 98 pips and reversed, you would end up with a BE or a Loss depending on your stop rules. Whereas a discretionary system would allow the trader to see price reversing and close out for a partial profit at, for instance, 85 pips.

So, you rule is TP at 1R; if floating profit reaches above 0.9R, set a trailing stop loss of 15 points.

The thing I hate about discretionary systems is that rules are made on the fly when the trade is running.

With DIBS, I strongly believe that even if we trade it with iron clad rules, it should be profitable. Just don't ask me for data to back it up coz I don't have any..
  • #9,472
  • Mar 26, 2010 9:20am Mar 26, 2010 9:20am
Quoting RussiaTrader
yes, but its a discretionary system and i stretch the rules due to my experience of trading with IBs.
the AUD trade had a pin bar first and then the IB.
in these cases i go long even when the day is negativ.
the trade is already closed break even.
Ah the pin-IB combination. Good call
  • #9,473
  • Mar 26, 2010 12:59pm Mar 26, 2010 12:59pm
you seem to be killing your own argument here. you hold up all of your rules as being valid, yet cannot get positive results.

how do you know that the "low liquidity" of mondays or fridays has the same impact on 4 hour as not trading before 6am gmt when trading on hour dibs. Maybe specific times of low liquidity have less of an impact the larger the time frame that you use?

have you tested the concept below on its own:

This one was one of MY filters. The idea is the market should first go into congestion and therefore an IB and then we trade the Breakout of the IB. If the IB is more than 70% (or any high number) of the prior bar, it does not really signify congestion but more of an indecision. So I stay away from it.

I am well aware of the modelling done by Nathan and posted by PC.
My backtesting was done on daily bars - 17 years of data with 4 different trailing stop losses.

Rules

  1. I take every IB i get as long as it is on the right side of my trailing stop.
  2. If I have an IB that forms on the MA i use for my trailing stop loss. I will use that IB to reenter if I get stopped out, otherwise I will not re enter an IB that I get stopped out of.
  3. I close half a position when a trade goes 1:1
  4. I close the entire position when I get a close beyond my moving average.

  • #9,474
  • Mar 26, 2010 1:01pm Mar 26, 2010 1:01pm
I've had some success using this method on a 4H chart and so I am liking it. I like to make things simple enough to work for me. This seems to fit. It seems to me we can only have successful trades when our position is with the price. A break out of an IB in the direction of the trend does this. PCs rules modified to 4H seems to work for me. Cut your losses short. Let your winners run. My only modification to PCs rules is that I move to BE as soon as I can after his first TP. PCs reminder about what Larry Livermore said "it was not the thinking...it was the waiting" has been very helpful.
Thank you to Dibs, james 16, Jacko and Bobokus
  • #9,475
  • Mar 26, 2010 5:02pm Mar 26, 2010 5:02pm
to the U/J long on the open of trading based on this IB that formed friday.....

still have 3 positions riding risk free
Attached Image (click to enlarge)
Click to Enlarge Name: usdyenlong4thpostion.jpg Size: 73 KB
  • #9,476
  • Mar 26, 2010 5:08pm Mar 26, 2010 5:08pm
putting on the third postion.

First position is risk free
Second position is not risk free yet......

will add the third position based on the IB that formed Friday
Attached Image (click to enlarge)
Click to Enlarge Name: 3rdguposition.jpg Size: 74 KB
  • #9,477
  • Mar 27, 2010 4:04pm Mar 27, 2010 4:04pm
Quoting razorboy
This one was one of MY filters. The idea is the market should first go into congestion and therefore an IB and then we trade the Breakout of the IB. If the IB is more than 70% (or any high number) of the prior bar, it does not really signify congestion but more of an indecision. So I stay away from it.
That is one of the final points which our friends here don't understand. They say PC didn't say this or that, but you must first try to follow the ones he said. Some of his statements would need many rereading to understand.

IBs are still a great way to trade much like pin bars and the rest. They require discretion all the time.

Kenn
Mystery Is The Final Product Of Ignorance. Hey! What's ma chart doin'
  • #9,478
  • Mar 27, 2010 5:10pm Mar 27, 2010 5:10pm
Quoting razorboy
you seem to be killing your own argument here....
Great approach! I'm still struggling to implement IBs on D1, cause in order to take original DIBS trades I have to wake up at 3:00am ... I'm too old to do it AND get to work....

Cheers!

HT
Many share my views with me. But I don't share them with them...
  • #9,479
  • Mar 28, 2010 6:21am Mar 28, 2010 6:21am
I suspect that you could use a period converter to start the trading day whenever you want and still play daily IB's and make money (a trend is a trend is a trend). You would end up with a different set of wins and losses (and different probabilities all together) but the larger the time frame, the less of an impact "the daily periods of lower liquidity" would have on you. Additionally, you would be looking at the market in ways most people wont be, which from what I have read, is a method used by some traders. Your stops and entries would not be grouped with everyone else's.

Having said that, there is only one way to test that hypothesis.......

alternatively, my daily bar trades often dont kick in between 6 pm and 12am Est, so you could still probably play this method if you input your trades when you wake up

Quoting HouseTrader
Great approach! I'm still struggling to implement IBs on D1, cause in order to take original DIBS trades I have to wake up at 3:00am ... I'm too old to do it AND get to work....

Cheers!

HT
  • #9,480
  • Mar 28, 2010 7:04pm Mar 28, 2010 7:04pm
sorry, I may have misunderstood you.......you NEED to play on daily chart cuz waking up at 3 am to play hourly inside bars is not compatible with a normal life.............found that out the hard way.

Its quite simple........just size down on the risk and have the patience to wait for an 8 to 10 K pip run.......its happened in the past and will happen in the future and don't bother reading the news or listening to the experts..the euro was supposed to hit 1.55 in dec/jan according to GS...........

I am sure that I could come up with filters to avoid certain trades, but based on the historical return, its not something I think I need to worry about

Quoting razorboy
I suspect that you could use a period converter to start the trading day whenever you want and still play daily IB's and make money (a trend is a trend is a trend)....
4 traders viewing now