ForexMorningTrade System

mjj said:
Smart Live Markets. I'm now using the Turbo EA...think there must be something wrong.
Click to expand...

Actually I've just looked at Marcs Turbo MT journal. There was no trade on turbo this morning because the ADR Filter was activated. Looks like it was a day I should have been using the original system :cry:
 
Monday: NT
Tuesday: Long TP +38 --> £1017 (+1.7%)
Today: Long SL -38 --> £979 (-2.1%)

can't seem to get clear of break even on my account....
 
Alpari (UK) 6.15 Standard Settings

Mon - NT
Tue - +35
Wed - -45

Am about to go into hibernation for August after tomorrow's trade.
 
July 27th
All in red today... Not even one pip in my favour, as soon as entered it went down:(

Fast FMT: -40

FMT: -40

EFMT: -38

Total: -77, -36, -90
 
Guys a few questions for you,

Has anybody considered running an optimization backtest? If so do you find it's improved your results?

Also, what period do you run your optimsation backtests for and how often? I feel this EA is very market dependant so am thinking a backtest from feb11 - july11.

I have 99.9% tick data so am interested in doing this.
 
leerees said:
Guys a few questions for you,

Has anybody considered running an optimization backtest? If so do you find it's improved your results?

Also, what period do you run your optimsation backtests for and how often? I feel this EA is very market dependant so am thinking a backtest from feb11 - july11.

I have 99.9% tick data so am interested in doing this.
Click to expand...


Many people LOVE doing back tests. They come up with different results every time, depending on the period, the broker, the spread, and at the end of the exercise you can come up with a really sophisticated preset, which might give you an exact starting time for every day of the week, depending whether the long candle filter is applied or not applied, the break even point, with add pips or subtract pips or trailing SL, the TP, and the SL, perhaps even separated between long trades and short trades (factoring in optionally whether big news is expected at 09:30 rather than 13:30 and whether or not it's employment news, house price news, interest rate news or just trivial stuff).

It's sometimes called curve fitting.

Whereas that will tell you what would have probably worked great in the past, it won't take account of future changes in market conditions, so applying such presets might not lead to a good experience next time round. Forex is a camel market, and it's not often productive to try and make it into a complex technical process which can be traded mechanically.

Thousands of back tests have indeed been done, but if you'd take my word for it, just accept that the recommended settings are performing better currently than previous settings, and result from meticulous analysis both by members and by the system producer.

But of course you can do your own backtests, February to July, or whatever period you like. There is no shortage of numbers and they might even contain next time's eurolottery winner! :cheesy:
 
wiseambitions said:
Many people LOVE doing back tests. They come up with different results every time, depending on the period, the broker, the spread, and at the end of the exercise you can come up with a really sophisticated preset, which might give you an exact starting time for every day of the week, depending whether the long candle filter is applied or not applied, the break even point, with add pips or subtract pips or trailing SL, the TP, and the SL, perhaps even separated between long trades and short trades (factoring in optionally whether big news is expected at 09:30 rather than 13:30 and whether or not it's employment news, house price news, interest rate news or just trivial stuff).

It's sometimes called curve fitting.

Whereas that will tell you what would have probably worked great in the past, it won't take account of future changes in market conditions, so applying such presets might not lead to a good experience next time round. Forex is a camel market, and it's not often productive to try and make it into a complex technical process which can be traded mechanically.

Thousands of back tests have indeed been done, but if you'd take my word for it, just accept that the recommended settings are performing better currently than previous settings, and result from meticulous analysis both by members and by the system producer.

But of course you can do your own backtests, February to July, or whatever period you like. There is no shortage of numbers and they might even contain next time's eurolottery winner! :cheesy:
Click to expand...

ah ! , Wise one, very well stated !!..(y)

there is nothing wrong with FMT or the default settings, but personally I have shut down FMT live trading last Monday until September 12. This week the market already looks like August market and is a dangerous time for most EAs. If you do decide to trade FMT during August I would suggest much smaller lots.

cheers !..:D
 
wiseambitions said:
Many people LOVE doing back tests. They come up with different results every time, depending on the period, the broker, the spread, and at the end of the exercise you can come up with a really sophisticated preset, which might give you an exact starting time for every day of the week, depending whether the long candle filter is applied or not applied, the break even point, with add pips or subtract pips or trailing SL, the TP, and the SL, perhaps even separated between long trades and short trades (factoring in optionally whether big news is expected at 09:30 rather than 13:30 and whether or not it's employment news, house price news, interest rate news or just trivial stuff).

It's sometimes called curve fitting.

Whereas that will tell you what would have probably worked great in the past, it won't take account of future changes in market conditions, so applying such presets might not lead to a good experience next time round. Forex is a camel market, and it's not often productive to try and make it into a complex technical process which can be traded mechanically.

Thousands of back tests have indeed been done, but if you'd take my word for it, just accept that the recommended settings are performing better currently than previous settings, and result from meticulous analysis both by members and by the system producer.

But of course you can do your own backtests, February to July, or whatever period you like. There is no shortage of numbers and they might even contain next time's eurolottery winner! :cheesy:
Click to expand...

The reason I backtest EA's is because in this industry there is no way to determine what standards the EA programmer has followed on his backtests.

Do you have any links to the backtests because I've looked and I can't find any?

We all have a rough understanding of the market dynamics of this strategy, so leaving out january would not be considered curve fitting as one wouldn't trade FMT in Jan anyway. The goal is not to produce an amazing backtest, instead it is to tune the EA into the current market using optimisation.

With regards to spread, you can set your spread to a fixed amount.

I am going to setup my own optimisation backtest and put my findings on a forward test, which I'll share on myfxbook.
 
You must log in or register to reply here.

Similar threads

W
29 turning 30. 11 years of trading. First profiting year Journal 1
2
Replies
37
Views
7K
IceThor
I
T
Colleged Fund Challenge
2
Replies
20
Views
4K
tkbieber
T
T
  • Poll
College Fund Challenge
32 33 34
Replies
678
Views
78K
tkbieber
T
Sharky
T2W Members' Choice Awards 2010 - The Results!
2 3 4
Replies
76
Views
19K
tar
tar
S
Question regarding high implied call volatility??
Replies
0
Views
2K
stocktradingguy
S
Top