Robust Volatility Breakouts - Set and Forget

hello guys, I'm very happy to find this thread. I like breakouts and this thread looks like very interesting

Asasa, your results are very impressive. Can you share your system rules?

Hei Transzorb, can you modify VALMAR indicator? instead pips buffer, a ATR buffer, for example, 30% ATR of previous day, or 10 days, etc

So it was a losing week. The number of pips is completely irrelevant; its all about $ risked versus $ returned. Be a trader, manage your risk, trust your strategy, and have the capital to come back and play next week and the next till you hit those weeks that pay you double what you risked. If you want to, backtest it yourself so you have the confidence to trade it forward.

hxcif do you have any plan on how to monitor these trades? Seems we would have to check the charts at least every hour. To catch the price movement when it hits the 1st green line so we can place the sl at breakeven.

thx,
todd

Herewith an updated version of the indicator with all the additions requested. For now we will leave it to this for the indicator. Next step is building a TradeStation program to do thorough back testing.

Regards,

T.

Kudos to everyone for all the work you've put in here.

I've actually been looking into a style similar to this lately. My staple used to be the London Breakout but I began testing different approaches to a weekly breakout such as the one being discussed in this thread but haven't gone live with it yet (I did run the daily ones live, just not weekly)

I just had one question about the point that tdion brought up:

While I would normally completely agree about the questionable quality of a backtest that's missing lower time frames, I was under the impression that the original idea of the First Strike method was (if I understood it correctly, and that's a big if right now) a single set and forget per side per week (one buy stop, and one sell stop, OCO style).

Assuming I understood that correctly, would it not be logically impossible for a lower timeframe to hit the stop and not the target (or vice versa) while a higher timeframe's Open/High/Low/Close data doesn't cross the stop loss price?

To rephrase that: If a lower timeframe (whether it's tick data or 1min to 1 hour) hits the stop loss, then the higher time frame (daily, weekly, monthly) should reflect that occurrence since the higher time frame's high/low value would have expanded to hit the stop as well. (The only exception would be if the market came within about 5 pips of the stop without hitting it, in which case it wouldn't matter what timeframe of data we use for backtests, most brokers would likely hit our stop in those cases anyway. But I don't think that's the issue at hand -- although it might be worth trying to count how many of the winning trades in a backtest came within such a distance?)

Sorry, back to the original issue:

Granted, this would be an entirely different issue if we were to assume that re-entries and such would be made, but I don't think that was the original idea.

So as long as the winning trades (ie. the runners that end up being closed on Friday after a move that might be 3x to 10x the side of our average losing trades) eclipse the size of our typical stop-out, then it should (in theory) be profitable. As I understand it, this method should allow us to do the reverse of the typical beginner habit (taking lots of small profits and then getting hit by one big losing trade that wipes out all the winners and more -- we would be doing the exact opposite. And in the few backtests I've done of this style, it seems that's exactly the case since there were streaks of 8 losers followed by one winning week that covered all of those losses at once.)

I don't mean to be blindly defending Joel's method, I've yet to even run a weekly breakout like this live... And I have as much skepticism as tdion about any system that comes from anyone who bait-and-switches a "free" system and then charges for seminars rather than just trading as a career.

I just started exploring this style because it's similar in concept to daily breakout strategies that worked for me, and a weekly scope would better fit my current schedule so I really appreciate the ideas and work everyone has put into this thread.

I'm just curious if I've misunderstood something about the concept of the method since lower time-frame data seems a bit irrelevant if we assume a "single entry and single exit per week" approach.

Transzorb thx for the update.

Looks like it's adding the spread fine on the long side, but it's also adding it to the sell side which I don't think it should be.

thx,
todd

Hi Todd96096'

Tranzorb and I developed the current version of the indicator. Can you explain why you think the spread shoud not be added to the sell side ?
The original code that was posted with the first message of this thread does the same (adding both to sell/long side?

thx
Esntrm

In addition to esntrm's comments, herewith a new version of the indicator that contains a small fix to cope with the difference between brokers with 4 digits price feeds and 5 digits price feeds. Beware that for Buffer and for Min_Range, if you want to enter 20 pips, in case of a 4 digit broker you should enter 20 and for a 5 digit broker you should enter 200.

Regards,

T.

thanks Transzorb, that's very kind of you.

I'm sure you already know why we want to add the spread to the buy side. But anytime we want to sell at a predermined line, there's no need to add spread. The sell will take place at that line. As apposed to buying at a line, you need to add the spread in order to have the trade take place at that original line.

just my opinion.

thx,
todd

i'm not having any trouble with my 5 digit broker using the 2nd version.

Hello,

Is anyone using this indicator for actual trading? If so, would you share how you're using it? Entry/Exit, S/L, T/P, etc.?

Thank you for sharing, hxcjf, and thank you for your code, Transzorb!

Thank you for your system rules, hxcjf. Can you post your trades, for this week?

hi hxcjf, I've found and old thread from you called "the lobster". Are you trading that system now? It looks like interesting.

So what happened to this thread, where did everyone go?

todd

Testing this week I got 80 pips on gj and 3 pips on uj with a loss on one side and tp on the other. Not trading the metals.

Hi HXCJF,

I've coded up the UJ system described in your DOC file, and I'm not seeing the same results as the "UJ weekly 1hr" tab in the XLS file you released. I'm backtesting with MT4 "history center" 1 minute data going back to 2006 with 90% modeling quality, and I'm using 3% account balance risk per trade. It moves SL to BE as prescribed, closes any open orders on Thursday at midnight ("after Wednesday"), and it closes any open trades on Friday afternoon.

What I believe I'm missing is the fact that sometimes you are taking 3x SL profit on the XLS sheet, whereas the DOC only mentions up to 2x SL as the TP level. Is each line a separate trade? (more than 52 lines per year, so I would assume so. Plus, the lack of any weeks with 2 losses)

Am I not understanding the "bracket" term correctly? I've attached an example pair of orders that the EA is setting up.

Also, what's the difference between the two "x gained" columns in the XLS?

I'd rather not post my code if it incorrectly implements your instructions so that people do not get confused.

Thanks!

FWIW, I attached my code's results from 2006.01.03 - 2010.12.06