Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Box Breakout indicator 195 replies

Breakout EA based on New York Box 3 replies

4H Weekly Box Breakout [TD] 4 replies

EURUSD Box Breakout 6 replies

Breakout Box 5 replies

Attachments: 4H Box Breakout
Exit Attachments

4H Box Breakout

Quoting ledobedo
i think the first week can be very profitable solely for the reason there are so many news. if u can anticipated the news, your intra day trading can turn very profitable.
I agree that the first week could be very profitable for intra day trading. ( I like trading in the first week) But this system is not an intra day system.
exactly, to be honest, in most cases, the news causes fakeout breakouts. Which is the all time murderer of this system.
ya i get what you are saying...you can always look at the calendar to filter these...maybe tigher stop losses in the first week could help
Quoting Ace284
No there is no flaw in your logic. I'm quite familiar with martingaling having read it quite a few times in the past. I know that it can get dangerous risking double, triple, quadruple...the amount of regular risk. The main thing we would need to look out for is whether it does get too out of control because that's where more martingale traders fail by actually blowing up their account faster.

2002 resulted in profit, have you checked the other years? Note down whether your positions got way too high. Looking to hear back.
The results with the semi-martingale (see post #2785 & #2790) from danno's download, as compared to Ace's method without the semi-martingale are as follows: [in pips]
2002: +1555 vs. -519
2003: +5303 vs. +3468
2005: +2685 vs. +49
2006: +2340 vs. +1496
2007: +4215 vs. +7474
2008: +7572 vs. +4135
2009: +7381 vs. +5028 (up thru 8/10/09)

I guess the results are self-evident...

Walt
Quoting jitasb
If a strategy didn't work 5 years ago, but is working for last 6 months, does that mean we shouldn't use it ? I was kind of thinking about a year seems a good period.
i think large period like 5 or more years can give a broad view of how strategy is working in different conditions which gives you better look at this system its weak points etc
but on the other hand as previously said this may have to do nothing with todays market behaviour, the question is to or not to use rules based for example on last 5-7 years , i think if it doesnt affect profits strongly it is a good idea becuase you never know when the conditions will change and your account especially when small is more secure
But i think it isn't also a bad idea just to make out rules from this year and just ride on wave being aware that conditions may change and when they will recognize this moment as early as possible and adapt. But with small money i would use safest option because one single week like this can kill small account.

my english... i think i make mistakes all the time but i dontt bother to look for them
Quoting jones247
The results with the semi-martingale (see post #2785 & #2790) from danno's download, as compared to Ace's method without the semi-martingale are as follows: [in pips]
2002: +1555 vs. -519
2003: +5303 vs. +3468
2005: +2685 vs. +49
2006: +2340 vs. +1496
2007: +4215 vs. +7474
2008: +7572 vs. +4135
2009: +7381 vs. +5028 (up thru 8/10/09)

I guess the results are self-evident...

Walt
I had a feeling that this would be the case using the semi martingale approach.Walt, the main question is, what is the average position size used for the martingaling?

Say in 2008 the average position size is twice as large using martingale (this is just an example). If that's the case, then you may as well double the size using the original strategy, that way you'll have 8000+ points.

Looking to hear back.
Quoting Ace284
I had a feeling that this would be the case using the semi martingale approach.Walt, the main question is, what is the average position size used for the martingaling?

Say in 2008 the average position size is twice as large using martingale (this is just an example). If that's the case, then you may as well double the size using the original strategy, that way you'll have 8000+ points.

Looking to hear back.
I would have to fully agree with Ace here, sure Martingaling shows positive results from 2000+, but what about pre 2000, when the average profitable breakout from the first 4H bar didn't occur until mid/late week, and the beginning was a savage whipsaw?
Interesting discussion going on about the amount of backtesting that's been done and the relevance of backtesting. (Thanks again, to Danno for his extensive help)

Guys to be honest, when I started trading this I didn't backtest for years. I noticed a pattern in the current market during last September until November and then jumped in. It was a three month observation in real time. Keep in mind that I just started trading back then, and didn't really see the point in testing extensively because I just wanted to make money

The main question with backtesting is...well is market behavior years ago really that important to me? Markets change, volume has definitely changed so should I really concern myself with 5 years ago.

Another good question is, how long in the past from today would you feel you need to backtest in order to trade this method?

These are certainly some of the most important types of questions for this type of method. Looking to hear everyone's opinion on this in the next few posts. Of course, I'll share mine as well in some time.

As traders, its important to take in all the information, sort out the relevant points and then making trading decisions off them.
Like a lot of you I use FXDD. I know how to place the opening orders, stops and targets but how do you move your stops to B.E. ? If I'm awake I know what to do, but how do you handle the "midnight movers" ?
Also, ACE, are you actually placing your orders thru FXDD? Do you have a broker you recommend (for trading, not just for charting) ?
Thanks
krly
Quoting Ace284
Interesting discussion going on about the amount of backtesting that's been done and the relevance of backtesting. (Thanks again, to Danno for his extensive help)

Guys to be honest, when I started trading this I didn't backtest for years. I noticed a pattern in the current market during last September until November and then jumped in. It was a three month observation in real time. Keep in mind that I just started trading back then, and didn't really see the point in testing extensively because I just wanted to make money

[b]The main question...
I agree, but the market has different characteristics within the last 5 - 10 years than it did say, 15 years ago.

I hate to bring it up, but the year 2001, the markets changed forever, I believe in doing long term extensive backtests because it shows if the system can handle catastrophic event that will ever change the market.

I believe the events of 2001 is what makes this system profitable post 2001, and murder pre 2000
Quoting yen44x
You have mentioned this several times. Do you have an Excel sheet or some sort of data that you can post to back this up? If it's true, it would be worth implementing, but so far you've furnished nothing to substantiate what you're saying.

If you recall, you made the same statement a while back, and I quickly scanned the charts and found 4 recent examples where a trader would have left hundreds of pips on the table by ignoring the second trade.

Yes I did. My computer/ Excel skills are limited and I cannot make Danno96 kind of spreads. But I already pointed out that you also made mistake regarding winning second trade while it was not.
Yes, there were few second winning trades but overall they did not compensate for other second trade losses.
1 week 2 losses is 2%. 2 such weeks is 4%. The vast majority of successful trades are the first trades of the week. You cannot argue about it.
Well, gotta go to prepare for my vacation
Keep your 1% intact
Quoting Ace284
I had a feeling that this would be the case using the semi martingale approach.Walt, the main question is, what is the average position size used for the martingaling?

Say in 2008 the average position size is twice as large using martingale (this is just an example). If that's the case, then you may as well double the size using the original strategy, that way you'll have 8000+ points.

Looking to hear back.
I used a scaled approach, starting with 1 lot. After two losses in a row, I would increase the size only by 1 lot. Also, everytime I had a successful trade, then I would scale back to the previous lot size. Typically, I would not go past 4 lots. On a few rare occassions, I would get as high as 6 lots. Since I only scale up after every two consecutive losses and scale down after each win, it keeps the semi-martingale from getting out of control. I think it would be too risky to scale up after each loss. Needless to say, that a "true martingale approach" (i.e. doubling up after each loss) would probably cause a blow-up of one's account.

Walt
Quoting shiko
Yes I did. My computer/ Excel skills are limited and I cannot make Danno96 kind of spreads.
Danno96 did all that manually. No skills are required to use Excel, all you need is patience. All you do is manually look at your charts, and put numbers into the boxes in excel.

Hmm, let's just say if you can post, then you can make an excel document. After a day, I think you'll be a natural

Have fun on your trip, we'll be looking out for some excel sheets when you're back.
Quoting shiko
Yes I did. My computer/ Excel skills are limited and I cannot make Danno96 kind of spreads. But I already pointed out that you also made mistake regarding winning second trade while it was not.
Yes, there were few second winning trades but overall they did not compensate for other second trade losses.
1 week 2 losses is 2%. 2 such weeks is 4%. The vast majority of successful trades are the first trades of the week. You cannot argue about it.
Well, gotta go to prepare for my vacation
Keep your 1% intact
I agree completely!!!! Using slightly different trade rules that Ace's, but essentially the same thing, the second trade lost at an extremely high rate.

Here are the results from my slightly modified version for 2009:

1st Trade of the week:
total: 35
wins: 15 (42.9%)
losses: 7 (20.0%)
b/e: 13 (37.1%)

2nd Trade of the week:
total: 18
wins: 3 (16.7%)
losses: 11 (61.1%)
b/e: 4 (22.0%)

Results such as this makes me wonder if it's best to only trade the 1st breakout and forget about trying to catch a breakout on the other side of the box. Also, perhaps to ensure that you're catching a legitimate breakout, it may be necessary to add more than 20 pips to the ends of the box, while setting a big SL to keep from being whipsawed (i.e. perhaps for small boxes, SL s/b at least 100 pips; for large boxes, SL s/b no more than 200 pips)

These are just ideas to continue the "think tank" spirit...

Walt
Quoting jones247
I agree completely!!!!...
Can you post the average 3rd breakout win rate?
With some people putting thier stop loss at breakeven they have risk nothing. Some move stop loss to break even after they reach 50%. Some after 100 pips.

How about entering the trade again after it goes bak in the channel to start again. I know we are only to do 2 trades. But you have loss nothing because you got out at break even. Just thinking outside the box.

Im using

hi-152.37
lo- 150.71
Quoting jones247
I agree completely!!!! Using slightly different trade rules that Ace's, but essentially the same thing, the second trade lost at an extremely high rate.

These are just ideas to continue the "think tank" spirit...

Walt
Walt and shiko . . . I did my own backtesting and I'm proud to announce that . . . you were both right and I was wrong.

My data was quite a bit different than yours, Walt, in regard to the weeks that started with a losing trade and then went into a second trade. That may be because I did not include weeks where the first trade went to at least TP1, since the rule now is that the trade would be moved to BE at that point. As a result, I found only 7 weeks this year where a true losing trade was followed by another trade. Of those 7, 1 second trade turned into a winner, 1 was BE, and the others were losses. The conclusion for 2009 up-to-date? You were both right and I was wrong.

Does that mean I'll stop taking second trades? Not just yet, and here's why: Remember, if only one of those losing trades had been a winner, the results of second trades would have been profitable for the year due to the great R:R of this system. I'm not sure I'm ready to draw any definite conclusions based on 7 trades.

Walt, if your data shows I'm wrong in my interpretation of what constituted a valid second trade (in other words, why you noted so many more second trades than I did), please let me know. In the meantime, I'm off for a second helping of humble pie.
Attached File(s)
File Type: xls 4H Box-Results of 2nd trade.xls   22 KB | 450 downloads
Greetings from Guanajuato, Mexico
Quoting shiko
Yes I did. My computer/ Excel skills are limited and I cannot make Danno96 kind of spreads. But I already pointed out that you also made mistake regarding winning second trade while it was not.
Actually, someone else pointed it out, but feel free to take credit for it.
Greetings from Guanajuato, Mexico
so the bottom line we should really only do one trade a week on this system
Quoting ledobedo
so the bottom line we should really only do one trade a week on this system
That may be your bottom line. It's your decision. As pointed out, though, due to the R:R of this system, if just one of those losing trades had turned into a winner, you would be in profit for the year. So you'll have to draw your own conclusions.
Greetings from Guanajuato, Mexico
1 trader viewing now