Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Trading the Sixths with Snowballs 68 replies

Gann High-Low System with Laguerre/Fisher/VSA/Sixths and others 124 replies

My Sixths trading journal with TOB EA 50 replies

sixths indicator 1 reply

Steve Trading Journal 3 replies

Attachments: Sixths trading - an EA by macman, Bob and Steve
Exit Attachments

Sixths trading - an EA by macman, Bob and Steve

I have added NB-style Recovery to TB - there are a few more details about this in post 1.

I am not sure the code is correct because the logic is complicated, for me at least.

Also, I am unsure what effect secondary trades would have on the process, so I suggest leaving them turned off for now.

I have extended the default MinPipsFromLastEntry to 50.

As soon as the demo send its first pending I realised there was a need to re-order some of the code.

Fix in post 1.

  • Post #63
  • Quote
  • Edited 11:53am Oct 7, 2010 11:37am | Edited 11:53am
Quoting SteveHopwood
As soon as the demo send its first pending I realised there was a need to re-order some of the code.

Fix in post 1.

Hi Steve,
My demo is up 600 points and running clean and smooth......
Question...I see a 0 on reentrylinepips does that mean that we are using Max's numbers and can I change it if I'd like?
Congratz....Looks like another barnburner.....

whip
Attached File(s)
File Type: pdf Statement_ 1158152 - Hiredwhip.pdf   82 KB | 431 downloads
Quoting hiredwhip
Hi Steve,
My demo is up 600 points and running clean and smooth......
Question...I see a 0 on reentrylinepips does that mean that we are using Max's numbers and can I change it if I'd like?
Congratz....Looks like another barnburner.....

whip
I will put a note up about this in page 1 later. Yes; set reentrylinepips to 0 and TB will use Max's table to calculate the value a la NB bot.

I have restarted the test with the latest version as I had accumulated a number of repeat trades.

But I am getting multiple limit orders. See attached example. I don't think I missed something but not sure. I have deleted the orders manualy.

Z
Attached Image (click to enlarge)
Click to Enlarge Name: TB GJ.JPG Size: 106 KB
Quoting Zr-fx
I have restarted the test with the latest version as I had accumulated a number of repeat trades.

But I am getting multiple limit orders. See attached example. I don't think I missed something but not sure. I have deleted the orders manualy.

Z
This is one of those mt4 is rubbish thingies. I will add a 2 minute sleep function after a successful trade-send to try to overcome this. If this does not work, I shall add a duplicate-trade deletion function.

Next on my immediate list is to add a Recovery break even profit function. I have benefitted from being able to use the break even line to add a few pips to the BE tp today.

In fact, I am trading TB live on GU, GJ and UC charts, so I wanted the function to add a few pips to the breakeven tp pdq. The update for this (with the added sleep thingy) is in post 1.

The relevant input is RecoveryBreakEvenProfitPips.

That is pretty much it for immediate development; now comes the fine-tuning and bug-squishing.

Steve, I don't think the recovery function is need or even beneficiary.

The fact that the lines are moving with PA over time you will have a kind of automatic recovery system already.
Try to backtest with recovery... and then with recovery off and secondary on... and you will see
"The only successful substitute for brains is silence."
Quoting DXTrade
Steve, I don't think the recovery function is need or even beneficiary.

The fact that the lines are moving with PA over time you will have a kind of automatic recovery system already.
Try to backtest with recovery... and then with recovery off and secondary on... and you will see
Yes, that is how it is designed, but a recovery system & good MPTM settings should reduce floating drawdown significantly - and profit slightly

Just a question of finding the right settings ....
Quoting SteveHopwood
This is one of those mt4 is rubbish thingies. I will add a 2 minute sleep function after a successful trade-send to try to overcome this. If this does not work, I shall add a duplicate-trade deletion function.

Next on my immediate list is to add a Recovery break even profit function. I have benefitted from being able to use the break even line to add a few pips to the BE tp today.

Generally I use :
Condition for open trade ......&& TimeCurrent()-LastTradeOpenTime>delaywait

when trade is send :
LastTradeOpenTime=TimeCurrent();
Quoting macman
Yes, that is how it is designed, but a recovery system & good MPTM settings should reduce floating drawdown significantly - and profit slightly

Just a question of finding the right settings ....
I back you to find them, too.

The problem of 'the runner' rears its ugly head again.

One of the many problems involved in backtesting in mt4's strategy tester is the way in which time screams by. Trades that go squillions of pips into dd happen so quickly. In reality, they take days, weeks, months to manifest themselves.

This gives us lots of time to decide what to do about the runners. Not that they happen very often.

Also, backtesting tests one pair. It ignores the potential of trading a lot of pairs and using the profit gained from profitable trading to offset the occasional runner against said profits.

And then there is hedging. I know that victims of US criminals are unable to hedge right now, and appear to be unable to do so for the forseeable future. Eventually, US residents will find a way around this. The rest of us merely have to transfer our accounts elsewhere.

So, hedging is a realistic defence against a really dramatic runner (assuming the losses from a runner are not easily offset against profits from the rest of the pairs).

The problem with hedging is when to start and when to close the hedge. Here Scoobs showed the way back in the BOJ thread; I have used his insight ever since.

What Scoobs said was this:

 

  1. Rsi on a D1 chart shows whether the market is trending or ranging.
  2. D1 Rsi >55: market is trending upwards
  3. D1 Rsi <45: market is trending downwards
  4. D1 Rsi in between these numbers: market is ranging

I have used this invaluable information ever since to take decisions to take/hold/close trades whether the trades were triggered by an ea or manually. I even have an ea that trades this information successfully on demo, but which needs refinement before it can be considered for general release. Well, what a surprise.

Drag the attached indi onto a chart - tf does not matter. It is a multi-timeframe Rsi. The TimeFrame of 1440 means the D1 tf. RSIperiod is the usual Period input that greets us when we drag a Relative Strength Indicator onto a chart. The indi that ships with mt4 defaults to 14 because this is the setting that Welles Wilder worked with for his own reasons due to the trading he was doing when he developed the indi. I use the attached default because it represents a month of 'real' trading today.

So, when deciding whether to hedge a short trade, this is the decision process:

 

  1. said trade is x pips in the hole, and has hit the should-it-be-hedged decision point
  2. D1 Rsi says the market is trending upwards, so the trade is hedged
  3. sometime down the line, the market returns to trending down as D1 Rsi demonstrates by falling below 45. At this point the hedge trade is closed.

All this is easily scaled out to a basket of Recovery trades that are deeply in the hole.

I do not plan doing anything about this for now because my experience of Recovery is excellent. It is simply something I am considering in the future.

Attached File(s)
File Type: mq4 #MTF_RSI.mq4   4 KB | 308 downloads
Quoting SteveHopwood
I back you to find them, too.

The problem of 'the runner' rears its ugly head again.

One of the many problems involved in backtesting in mt4's strategy tester is the way in which time screams by. Trades that go squillions of pips into dd happen so quickly. In reality, they take days, weeks, months to manifest themselves.

This gives us lots of time to decide what to do about the runners. Not that they happen very often.

Also, backtesting tests one pair. It ignores the potential of trading a lot of pairs and using the profit gained from...
There could be an interesting excercise todo here. Bobs system(s) use the theory of counter trend trading. Now, you may find that you get far less -DD or runners if you wait until the Daily RSI is above 70 or below 30 before firing retrace trades off. The thinking behind this is that the markets will be exhausted in either direction and are much more likely to retrace.

So, if your existing system rules say enter a sell then makesure the Daily RSI > 70 and if a buy makesure the Daily RSI < 30.

Ofocurse this is likely to mean overall allot less trades but you may find you can go for larger TP to coneract this in someway.

Its a great shame that MT4 brokers dont allow you to purchase bonds as this is the only real way to hedge as apposed to hedging the same symbol which can get you into a nightmare scenario.

Scoobs.
Quoting scooby-doo
So, if your existing system rules say enter a sell then makesure the Daily RSI > 70 and if a buy makesure the Daily RSI < 30.

Ofocurse this is likely to mean overall allot less trades but you may find you can go for larger TP to coneract this in someway.


Scoobs.
Nice idea again scoobs. Easy to code so I am about to do so. Cheers.

Quoting SteveHopwood
Nice idea again scoobs. Easy to code so I am about to do so. Cheers.

Some general rules regarding Daily RSI levels...

Above 70 = Overbought.
Between 55 and 70 = trending up.
Between 45 and 30 = trending down.
Below 30 = Oversold.

This doesnt work very well on lower TF than D1 because the commercial/big boys dont generally trade lower than the Daily. When a currency goes below 30 or above 70 for no reason then the large financial traders will load up trades. This is why you usually see some big reversals at these levels.

But please dont try this on M1 otherwise you will end up with fcuking huge -DD. And dont make this your only criteria for entry, it should be used as a final decision maker not a primary one.

Scoobs.
Steve,

Here you can see on the daily what happens when price touches 30 or 70.

Generally, fcuking huge reversals.

Scoobs
Attached Image (click to enlarge)
Click to Enlarge Name: daily7030.gif Size: 36 KB
Here is a version of TB with a Rsi D1 filter.

It appears to kill trading stone dead, but feel free to play with it and see if you guys can make it work.

This is not surprising. It takes a looooooooonnnnnnnnnnnnggggggggg time for the D1 Rsi to go overbought/sold; this will cause us to miss a lot of TB trades along the way.

Not a criticism, scoobs; merely a reflection of the reality of waiting for the D1 Rsi to go overbought/sold.

Attached File(s)
File Type: mq4 The Beast with Rsi.mq4   57 KB | 330 downloads
Quoting SteveHopwood
Here is a version of TB with a Rsi D1 filter.

It appears to kill trading stone dead, but feel free to play with it and see if you guys can make it work.

This is not surprising. It takes a looooooooonnnnnnnnnnnnggggggggg time for the D1 Rsi to go overbought/sold; this will cause us to miss a lot of TB trades along the way.

Not a criticism, scoobs; merely a reflection of the reality of waiting for the D1 Rsi to go overbought/sold.

Yes this is the drawback for retail traders, they wont wait for the cross overs. But you can load up on the lot size. Also, as a test you might want to set the Daily RSI Period to 7 instead of 14.

Scoobs.
Here's 2 days(+or- a few minutes) OOTB....H1...10 charts.
Very Nice....

whip

I do not have the TB with RSI update in yet....Steve, this thing has you and Macman written all over it...
Attached File(s)
File Type: pdf Statement_ 1158152 - Hiredwhip.pdf   83 KB | 518 downloads
Quoting hiredwhip
Here's 2 days(+or- a few minutes) OOTB....H1...10 charts.
Very Nice....

whip

I do not have the TB with RSI update in yet....Steve, this thing has you and Macman written all over it...
Lovely.

The version with Rsi is for backtesters to play with, so I would not rush to put it on demo just yet.

Steve,

You could look at using the RSI filter in a different manner so that you still get lots of trades with the original system.

i.e. Makesure you are not selling into oversold level and not buying into overbought level. So if NB or whatever says buy then makesure the daily rsi is not > 70 and if sell then maksure the daily rsi is not < 30 etc.

Scoobs.
0 traders viewing now