Strategy Setup for MTDoubleShotUSARTrader

Originally from ticket #3362.

I back tested a sample pair with the default settings on MTDoubleShotUSARTrader. I tried different dates and time periods. The results were not impressive. In fact, in many cases, there were no winning trades at all. So I assume the default settings are not optimal.

My goal is to trade 5-6 pairs at a time using mini-lots, with a total profit target of $100 a day minimum, and an average-risk on draw-downs. What settings would you recommend? With what pairs, time of day, and periods? Should the strategy be used in conjunction with Easy Trader, and how is that done?

I can understand some of the setting meanings from the gray box, and from introductory videos. But others are not so clear. So any general definitions or guidelines would be useful.

For optimization testing, what parameters would be the best to vary?

Although this is a self-contained strategy, can it be used with other strategies and indicators? If so, which do you recommend, and how would that be done?

Are there components of the strategy that can be disabled? Can that be done with settings, or must one go into the script to comment-out (disable) components?

*************************************************************************************************************************************************************************************************************

Sounds like Forex Pairs

We can legally offer no trading or investment advice only how to use features of the software and an approach.

Basic Settings are provided to show features - You get a suit of the peg and now you need to make it fit.... you only know what will fit you. everybody is different.
This is 1000,000 % true - could you just switch on with someones settings and stick with it that make money for them>? 99% no... The Turtles told people their algo and didnt care as they knew 90% would not be able to follow it.... the answer is you would unplug it.....

You need hands on experience of that process to create enough belief in time say 1 to 3 months of study to be able to stick with a system and not unplug on the first drawdown....
Historical Data: First thing to do is procure 3 months to 3 years of data or more. i like 90 days to 180 days for my charts so i can get a feel for short and intermediate intraday views and i go back to 1995 on the higher time frames - as a commercial would do...

New Setting... but saying that i have some advancements in the trade signals it gives now based on feedback from a fund that use this to swing trade a basket of futures.... and these guy make money from trading from a living so i am hopeful that will work with forex... when the middle line (atrmedian) goes red and green.... NT7 i have added a new trade entry and will ship that out to you sometime around monday - there will be a new update.

"My goal is to trade 5-6 pairs at a time using mini-lots" -so make sure you chose the majors with the best ranges and follow through for a trend model... for backtesting you might want to focus on
USD pairs due to way NT reports $ using them.... but realtime will really only show you the possibility ... diversify also will help...

Setup mutliple sim accounts and run realtime variances of tests on the same instrument and track the realtime stats for specific settings/models - "as a pro fund would do with live cash or sim"

" with a total profit target of $100 a day minimum"
why limit the system - let it run and take out profits at various points to stop a retracement giving ti all back. take an average over the week/month/ year to see that figure.... historically you need the same conditions in the future to achieve that... what are those conditions? ranges? trends?

" and an average-risk on draw-downs. " only you can decide what that means - what is profit/maxdrawdown, how long it takes to recover from a drawdown, how many winnin days versus losing - expectancy... what is the average trade size?

"What settings would you recommend?"
we cannot recommend any.... but we can tell you how to discover "yours"

"plain vanilla mode"
start with none and add - ie take off all trade management ,stops, targets, exits, allow 24/7 trading -
allow dataseries to guide you..... small 1000s of signals to say 1 to 5 signals per month.... zoom in and out....

see the true system profile....and stats... and use one as your base model....

now.... consider - stops.... walk in a large stop from 100pips to 15 pipss... where is it best 25 to 35? can you trade that> can your account? dont let the fear of a small stop prevent you tradinng a winning style - swing...for example... if intraday is no good for this system...or perilous....

" With what pairs, time of day, and periods? "
dataseries will be the most important decision and how to trade... swing trading is much better and easier for this type of model.. consider if you want to stay in over the weekend.. and market exposure...also...
remember the Swiss franc!!!

try 1000 volume bars 10,000 vol bars for (futures)
or tick for example
100, 150, 250, 500, 1000, 5000
10 range, 20, 30 range
5, 10, 15Minute, 30, 60, 120

FOREX TOOLS for intel.... If you install the MicroTrends NinjaTrader Framework with all workspaces> Open up a workspace called "MT FX Multi Time Frame Analysis"
First: Download 10 years daily for all FX, 90 to 180 days minute , tick (all last price)
http://screencast.com/t/7vp2Epdb

This will show you the best ranging pairs... it will also allow you to spot oppurtunities based on price analysis/ price cycles and look to trade the breakout of congestion or bounce or fakeout and contintuation of levels using a commercial view with intraday entry timing... you can use the intel for when its best to use a system perhaps..if you can see it that way. you can see the

Should the strategy be used in conjunction with Easy Trader, and how is that done?"
MTDS is 100% seperate to that and not used at all - you have an equivalent called the NinjaBuddy - please see this video on how to use it:
http://www.youtube.com/playlist?list=PLtKS2GMs9w4U5sfvbaHt8Flv99oVLBMRP
and
https://www.youtube.com/playlist?list=PLtKS2GMs9w4WD6jg0FIpvpEjryKToSNfN (renamed there as woodies chart trader)

Simple Steps
To begin with - keep it simple.... do not use all the myriad of different features at the begin....
Just let it stop and reverse to begin with to see the true profile of the system on a dataseries
So then experiment with dataseries

Then configure that as a base plan - run a backtest visaully study on a chart what the parameters do - by changing them - do they make sense could you really trade that?
What can be implemented to reduce the number of trades... is it better for swing or intraday ( swing is easier) - reduce losers - with filters or different entry methods....
but you will also filter out winners... you got to be in the real big moves and let it run... yes the cliches are 100% true with this type of system...

Trend systems win: 35 to 40% win ratio and a positive risk reward 1:1.5+ is what you want to aim for - all the other stats will fall out of that

"For optimization testing, what parameters would be the best to vary?"
You can use the above test that as the base for a strategy analyzer test and optimize the dataseries - this is the most important decision...
and you need a clear view of the purse signals for a stop and reverse only setting to understand the potential - So take off all stops, be triggers, trail sstops and targets and exits and filters and then add some in gradually to see the affect -
Then add them in... first of all a stop... then maybe a trail - decide to stay in or go flat friday.... until you got something that seems to work for you....

Two models in parrallel and change difference groups...
use two models the same at the start
change one with trade management only
chance one with filters...
then see how to combine if needed

SELECTION TESTING - walk forwards
then run your selection in realtime

Realtime Sim Models -
use fully auto sim models with wide open settings to see the MAE and MFE and true drawdowns on the different price cycles
use that info to guide you when auo trade launch and when to exercise discretion, based on price cycles and stats... and look to the higher time frame to see when its more probable a trend is going to start...and give great follow through...

Think about the modes of operation.
How do professionals trade? A:100% fully? b:or do they use experience over the mechanical - c: do they add manual trades....and interact
B and C is more likely - using a realtime trading chart contrl such as the Ninjabuddy...to pause autotrading, filter long or short only, manually add trades or control the trade management....

Process
This process can take 1 day to 3 years i have been told by 1 trader....for people to accomplish or never - so hence its free.
for your experiment with this i would say take 1 to 3 months - 1 week to 2 weeks base plan discovery - 2 weeks filter testing 1 to 2 months realtime trading on sim....

I hope some of that made sense to you?
Its better you have to work hard and learn how to trade/an approach you can use for any system and really stand a chance of reaching your goals - versus me saying i use this try that.... trust me... that wont work....

i look forward to getting you an update....

P.S: NT8 - running the USAR http://screencast.com/t/TvLPxYnXS9Wh

related subjects/systems:
Session Open Breakout robot with USAR Trail
http://screencast.com/t/7Eu3s4S8
i will also be releasing that as part of the MTDS - i like that for forex....it seems to be potentuially viable will assist you with a test of that.... and try to put those settings in for you to "assess" as a guide to the features....

Tasks for ME:
so new release pending for you.... will add some templates to demo some of the above i discussed... ie plain vanilla mode

Tom

Disclaimer
Trading and investment carry a high level of risk, and MicroTrends Ltd does not make any recommendations for buying or selling any financial instruments. We offer educational information on ways to use our sophisticated MicroTrends trading tools, but it is up to our customers and other readers to make their own trading and investment decisions or to consult with a registered investment advisor.

MicroTrends Support
support@microtrends.co
http://www.microtrends.co/support/

Tip! You can visit the help desk forum to see previous answered frequently asked questions:
http://microtrends.zendesk.com/forums