Calling all swing/position/trend traders

[Rooster34]

Ok, I work long hours in a crappy time zone to trade in. Mountain standard time. GMT -7.
I'm starting to look at the longer term. 4hr/daily/weekly.
I need to be able to come home at night check the charts and set my orders and not worry about them untill I get home the next night.
Any suggestions, techniques, or recomendtions would be greatly appreciated.
Royce.

 

[meanreversion]

Trend following? Once you have your system sorted, it will require no more than 30 minutes a day to manage. Each morning, you will update your orders and then wait.

 

[Rugby Trader]

yep, i agree that's what i'm about to do. been paper-trading a trend following strategy (profitably thankfully) for 14 weeks and about to go live. Only takes 5 mins in the morning and I don't look at it again till the next morning where I close my previous positions and set new orders.

 

[Black Swan]

Any suggestions, techniques, or recommendations would be greatly appreciated.
Royce.

Yep, big stops are nec. and patience, you can be in a trade for days before it shows any progress. Perhaps alter your mind set; you're taking your current units of currency and exchanging them for others as a short term investment.

In terms of stops, if you're on dailys, perhaps look for the low or high of the previous 2 days as your stop. This can easily be 250-300 pips as an average (for me anyhow).

 

[meanreversion]

As Black Swan says, patience is required. Also fortitude, in order to withstand those big swings in equity.

Not to put your system down or anything, but the last quarter has been a good one for trend systems in general (i.e. USD lower).. have you backtested your strategy over a decent length of time?

 

[Rugby Trader]

Is 14 weeks enough? I've been forward testing and my paper profits on EUR/USD have been good, but I'm well aware that there has been a major trend recently.

 

[HowardCohodas]

Is 14 weeks enough? I've been forward testing and my paper profits on EUR/USD have been good, but I'm well aware that there has been a major trend recently.

Paper trading has it's limitations at mimicking reality. Before you qualify your strategy for full production you need to trade it for small money to overcome those limitations.

 

[Rugby Trader]

Paper trading has it's limitations at mimicking reality. Before you qualify your strategy for full production you need to trade it for small money to overcome those limitations.

I'm going to be going live with 10p stakes on smart live markets as soon as I money is transferred over from another account. So I guess I'll find out whether my method really works or not!

 

[Rossini]

how long have you back tested for?

 

[meanreversion]

Is 14 weeks enough? I've been forward testing and my paper profits on EUR/USD have been good, but I'm well aware that there has been a major trend recently.

I think by this you mean you haven't backtested - you mean that you've traded on demo for 14 weeks?

If your system is purely rules based, you should aim to backtest for at least 5 years, i.e. code it up in software and then run it on historical data. If you can go back 10 years, that's even better.

If your trading is discretionary, then you'll just have to trade it in the market and learn over time. You should get a good idea after a year or two if it's working for you.

 

[Shanghai]

If your trading is discretionary, then you'll just have to trade it in the market and learn over time. You should get a good idea after a year or two if it's working for you.

I wouldn't necessarily agree with that. Computer backtesting has its place but I like backtesting the old way from a chart by hand, tabbing forward one bar at a time. Takes longer of course but it does expose you to some of the real life dilemas of trading and decision making that computerised backtesting can't. An added bonus is that discretionay trades can also be backtested that way.

 

[meanreversion]

I wouldn't necessarily agree with that. Computer backtesting has its place but I like backtesting the old way from a chart by hand, tabbing forward one bar at a time. Takes longer of course but it does expose you to some of the real life dilemas of trading and decision making that computerised backtesting can't. An added bonus is that discretionay trades can also be backtested that way.

That sounds dangerously like effort!

 

[Rugby Trader]

I think by this you mean you haven't backtested - you mean that you've traded on demo for 14 weeks?

If your system is purely rules based, you should aim to backtest for at least 5 years, i.e. code it up in software and then run it on historical data. If you can go back 10 years, that's even better.

If your trading is discretionary, then you'll just have to trade it in the market and learn over time. You should get a good idea after a year or two if it's working for you.

yes i've been demo trading for 14 weeks.
i do have a rule set but whether the answer to each rule is yes or no is discretionary so couldn't be programmed unfortunately.

 

[Rugby Trader]

I wouldn't necessarily agree with that. Computer backtesting has its place but I like backtesting the old way from a chart by hand, tabbing forward one bar at a time. Takes longer of course but it does expose you to some of the real life dilemas of trading and decision making that computerised backtesting can't. An added bonus is that discretionay trades can also be backtested that way.

I would quite happily spend plenty of time doing that. Not sure my girlfriend would be so happy though.
Where could I get the charts from?
I need 4hr and 1hr charts to check my rules and then 1min or 5 min charts i suppose to check when entry and exit rules were hit.
i currently just use spreadbet providers so the charts don't go back far at all.
if i signed up with a proper broker like alpari would they supply them?
and would i be able to put a simple moving average on those charts, that's the only
indicator i use.

 

[meanreversion]

Then Shanghai's suggestion is a good one. It really will be worth your time and effort to attempt something along these lines. Although it's not often stated, the psychology of trading is sometimes more important than the system. Most new traders focus on finding a perfect system, but even if you do find something that works for you (whether mechanical or discretionary), there will always be times when the system is losing money (by definition). You need to build confidence in what you're doing, in order to stick with it through the inevitable drawdown.

You'll notice that very few people on this forum use the term drawdown.. most of the comments are along the lines of "I make 20% a week" or "I'm up 100% this month" or even "I only enter trades with good risk/reward".

You can only set the "risk", the market will decide your "reward". Think less about how much you want to make, think more about how much you're prepared to lose, in order to generate a certain magnitude of return.

 

[Black Swan]

I wouldn't necessarily agree with that. Computer backtesting has its place but I like backtesting the old way from a chart by hand, tabbing forward one bar at a time. Takes longer of course but it does expose you to some of the real life dilemas of trading and decision making that computerised backtesting can't. An added bonus is that discretionay trades can also be backtested that way.

Agreed, and if you're honest with yourself ( and are trading on the 4hr/dailys) you should have at least your 6 month *backtest* results staring you in the face from day one...Curve fitting? Only the trader/user of the strat. can decide that one...

The only other thing I'd add is that if you're going to swing you have to be in several currencies, not from day one obviously, you should simply scale it in when your alert/s tell you too. But IMHO you can't win at swinging unless you're into all the majors...

 

[meanreversion]

The only other thing I'd add is that if you're going to swing you have to be in several currencies, not from day one obviously, you should simply scale it in when your alert/s tell you too. But IMHO you can't win at swinging unless you're into all the majors...

Maybe Toast can suggest swinging to his missus.

 

[Rugby Trader]

I'm demo-ing 8 different pairs at the moment, AUD/USD, EUR/GBP, EUR/JPY, EUR/USD, GBP/USD, USDCAD, USD/CHF and USD/JPY.

Could anyone point me in the direction of where I could get the charts too backtest. I would really like to backtest my method on each of those pairs for a couple of years each.

 

[meanreversion]

ProRealTime is an excellent charting/backtesting package, for the FX package I think it's about EUR 400 a year, and their data goes back a loooong way.

Be careful about trying to source free information on the web as much of it is of dubious quality.

 

[meanreversion]

Sorry, I should mention that if you open an account with IG, you get access to the ProRealTime package. If you trade a certain number of times a month it's free, otherwise I think they charge you £30 or so. However, the version that IG offer is nowhere near as comprehensive as that which you get directly from the vendor.