Inside Bar Trading Revisited | Page 37 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

3 Bar Fractal with mid-bar and bar on each side 11 replies

Inside bar or outside bar identifier 7 replies

Journal related to: Advanced Inside Bar (IB) Trading 19 replies

Pin bar, pin bar, pin bar 0 replies

Attachments: Inside Bar Trading Revisited
Exit Attachments

Inside Bar Trading Revisited

  • #721
  • Mar 16, 2009 10:51am Mar 16, 2009 10:51am
Quoting piprac
... on Candlesticker.com an IB (harami) has low reliability for reversal. An OB (engulfing) has medium reliability.
Thanks piprac. Interesting to know. The EA results I post here are all based on OB and generally run with default values. I think the currency based hourly charts are the most useful, as these can give you a good starting point for further optimisations.

However Abnormalone's point on brokers is valid and the results I post for Alpari may not be the same as other brokers.
  • #722
  • Mar 16, 2009 11:58am Mar 16, 2009 11:58am
Quoting piprac
I did some reading & looking over old notes & strategies this past weekend and on Candlesticker.com an IB (harami) has low reliability for reversal. An OB (engulfing) has medium reliability. I wonder how much chart watching & EA effort should be devoted to low reliability? At the least perhaps one should consider only OBs. And I don't know how accurate or tested "candlesticker's" methodology was to determine "low" or "medium" reliability. Comments anyone?
Hi piprac et al, hope you don't mind me chiming in here.... I trade ibs in conjunction with s&r, momentum etc.

I saw what you said "on Candlesticker.com an IB (harami) has low reliability for reversal" so the reverse would be true (they had a high reliability for trend continuance)? as that is basically how i trade them.

I'm not sure this method would transfer to an ea very well? I wrote one (i'm not the best but i try lol) that used ma's to filter trend which worked nicely but most failed trades were due to the ib falling near (support/resistance) I didn't and still don't know of any ea that has the ability to interpret charts in this way....

kind regards
  • #723
  • Mar 16, 2009 12:12pm Mar 16, 2009 12:12pm
I'm not sure if the reverse is true; "Candlesticker" gives 3 high reliability bullish patterns & one bearish. There are also mediums & lows shown too.
  • #724
  • Mar 16, 2009 12:29pm Mar 16, 2009 12:29pm
Quoting piprac
I'm not sure if the reverse is true; "Candlesticker" gives 3 high reliability bullish patterns & one bearish. There are also mediums & lows shown too.
i'm not familiar with "candlesticker" .... just had a look, it's all in the definition of IB's, "harami" "harami cross" "homeing pigeon" and "matching low" are all ib's but they study the price action within the bars context, where as I'd just look to see if the bar was within the previouse bars h/l (price action of the bar wouldn't have the same significance with ib's)

so apples and oranges really.

regards

intresting site though
  • #725
  • Mar 16, 2009 12:52pm Mar 16, 2009 12:52pm
iam following thise thread and anjoy
just whant to ask is it "sefe" to trade thise metod on the 15 m 30 m h1?
what the avg win/lose on it?
what the best way to trade tise metod?
i read same of the dibs thread i think they heve there same usful stuf
did u guy discove any new way to trade it?

i appreciate all u r work guys

thanks to 1ka and all of u

superscalper

  • #726
  • Mar 18, 2009 5:41am Mar 18, 2009 5:41am
1ka,

1. How does the EA handle requotes?
For example, if you are in a losing trade and it needs to SL but gets requotes. Will it keep trying to SL with the requoted price or leave your trade open (OUCH)?
2. Could you also comment on how Maxstart works with some numbers?
  • #727
  • Mar 18, 2009 10:08am Mar 18, 2009 10:08am
Quoting kostas1
1ka,

1. How does the EA handle requotes?
For example, if you are in a losing trade and it needs to SL but gets requotes. Will it keep trying to SL with the requoted price or leave your trade open (OUCH)?
2. Could you also comment on how Maxstart works with some numbers?

The SL is set on the trade, the broker software should take care of this automatically and requotes shouldn't be a problem as we're not actually asking the server to SL dynamically.

As for other things, open/close/modify, then the EA does not requote and we miss the trade. In theory the next tick should produce a new request.

The MaxStart attempts to stop a trade opening if it has moved past the Outer Bar. So if MaxStart is 100 (10 pips with a 5 digit broker) then the price at which a trade is opened must be within 10 pips of the Outer Bar. It also avoid problems when we add the EA to a chart and it wants to open a trade straightaway - the price may have moved significantly past the OB point.

  • #728
  • Mar 18, 2009 3:29pm Mar 18, 2009 3:29pm
Quoting 1ka
The MaxStart attempts to stop a trade opening if it has moved past the Outer Bar. So if MaxStart is 100 (10 pips with a 5 digit broker) then the price at which a trade is opened must be within 10 pips of the Outer Bar. It also avoid problems when we add the EA to a chart and it wants to open a trade straightaway - the price may have moved significantly past the OB point.


Thanks!
  • #729
  • Mar 21, 2009 8:21am Mar 21, 2009 8:21am
Cumulative trading charts. Last weeks charts are here.

0900 GMT still seems like a good time to trade, as does 1800GMT. EURJPY still the big loser overall.

Note that I've stopped posting the daily charts. Since the EA is running without any optimisation then the daily results are not really meaningful. I might substitute some of the live results instead.
Attached Images
  • #730
  • Mar 21, 2009 8:29am Mar 21, 2009 8:29am
Attached are the individual currency charts showing when it's a good time to trade. Note that these results are from Alpari UK's demo feed and the hours are GMT+1 on the 30M charts using Outer bar IBs. The results date from late Jan. The EA has been configured with default settings and has no optimisations applied.

The previous charts are here.
Attached Images
  • #731
  • Mar 21, 2009 8:30am Mar 21, 2009 8:30am
...and the rest.
Attached Images
  • #732
  • Mar 21, 2009 9:14am Mar 21, 2009 9:14am
I'm contemplating changing the individual currency charts to show the daily/hourly breakdown as shown in the attached images.

However, since the number of trades per hour is comparatively low, then statistically the results are a little meaningless.

More useful or less useful?
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Picture 1.png Size: 37 KB
Click to Enlarge Name: Picture 2.png Size: 37 KB
  • #733
  • Mar 29, 2009 4:40am Mar 29, 2009 4:40am
Cumulative trading charts. Last weeks charts are here.

0900 GMT still seems like a good time to trade, as does 1800GMT. EURJPY still the big loser overall.
Attached Images
  • #734
  • Mar 29, 2009 4:45am Mar 29, 2009 4:45am
Attached are the individual currency charts showing when it's a good time to trade. Note that these results are from Alpari UK's demo feed and the hours are GMT+1 on the 30M charts using Outer bar IBs. The results date from late Jan. The EA has been configured with default settings and has no optimisations applied.

The previous charts are here.
Attached Images
  • #735
  • Mar 29, 2009 4:46am Mar 29, 2009 4:46am
...and the rest...
Attached Images
  • #736
  • Mar 30, 2009 12:19am Mar 30, 2009 12:19am
1ka,
I am playing with the EA on a demo for a while. It seems that an alert with the description "IA detected: Will trade short at X.XXXX EURUSD" or something similar will be useful in notifying us that a trade is coming.
Again not necessary but useful....
  • #737
  • Mar 30, 2009 3:45am Mar 30, 2009 3:45am
One more idea (weekend was very rejuvenating....): What about a head-fake feature, i.e. a switch that would not trigger a trade the first time the OB is crossed but the second? It may save us some fake breakouts.
  • #738
  • Mar 30, 2009 3:53am Mar 30, 2009 3:53am
I trade some IB's... But, the only direction i go is with the trend, I only trade during high volume, and I watch the tape.

If you don't get particular on the IB's you take, then you are likely to get bashed in a tight range market.

Daily IB's are the safest.
  • #739
  • Mar 31, 2009 4:44am Mar 31, 2009 4:44am
Thanks Takisd and kostas1 for the comments.

I might try some of the higher timeframes at a later date. I know Abnormalone had some succcess with 4H+ (though my memory could be faulty ). I'm committed to 30M at the moment as I have a backlog of historical data to support that timeframe.

I like the idea of taking the second, or subsequent, breakout. If I get time in the next few days I might try that to see what impact it has.

I didn't bother with an alert feature. There are a number of indicators for IBs out there that already do this and I use them in conjunction with the IB. See here for an example.
  • #740
  • Mar 31, 2009 9:01am Mar 31, 2009 9:01am
Quoting 1ka
Thanks Takisd and kostas1 for the comments.

I might try some of the higher timeframes at a later date. I know Abnormalone had some succcess with 4H+ (though my memory could be faulty ). I'm committed to 30M at the moment as I have a backlog of historical data to support that timeframe.
So far i have not been able to find any settings for higher timeframes that will repeat when I shift from one time period to another. My best success has been mostly 30 min. and 1 pair on 60 min.. When I optimize for 1/1/2009 thru 2/28/2009 I have found a few pair that I can then go back or forward in time and still get good results. I am forward testing E/U E/G E/J G/U as they seem to preform best on IBFX. Still working on FXDD but not getting good results so far.
0 traders viewing now