Secret Bob. Sometimes trendy, sometimes not.

[Pipstubborn]

Hi Rapple,
I think it was deleted with the new Bob's update. Can you re-post it?

Thanks

Hi Pipstubborn

I used this one. Just note that you have to set Open Orders and Closed Orders to True/False depending on what you want to look at. I have both set to True so I can determine overall Profit/Loss

Kind regards Rapple

Thanks Rapple,
Will loaded tonight.

I notice that some guys are not very happy with the results of Mon, Tue. IMHO I think that we can't cut all the cookies with the same cookiecutter. On Friday all the pairs were pretty flat, therefore good reslults.

Through the D1 CSS reads I separate the pairs in two groups, Trend and Range pairs and applied the EA only to the ranging pairs and with enough distance between the arrows. Starting tonight I will separate the ranging pairs with the H4 CSS reads and apply an 80-100% ATR in an attempt to TP profit before the opposite arrow closes them.

A good example will be the EU/US that was ranging pretty well from Friday to halfway Monday. On the second half of Monday till now is down trending according to the H4 CSS and the H4 240 MA so I use the EA H4 240 MA filtered or Best Bob for trending pairs.

As Steve says an EA is not a set and forget but makes life easier besides that we need to keep in mind that is been developed and is too early.

Hope this helps,
Pipstubborn

[ianp]

At the risk of getting Steve's back up ,

I was wondering if the EA could be made to run on Renko charts ? I have no idea if it would work any better than time charts , but I thought If it was not to hard to adapt the EA it would be worth a forward test ?

I came across this thread http://www.stevehopwoodforex.com/phpBB3 ... 59#p102159

[Geges]

Just notice a phenomenon that when I change time frame from H1 to H4 then back to H1, some of the arrows disappeared upon which position has been entered based on the signal, particularly if you are trading multiple positions. Therefore when you eye ball the chart history, it looks much better than reality.

See attached charts which is taken just before and after a time frame change. The first, third and fourth arrows disappeared. I am trading two positions and the first and second arrows have been acted upon with long positions.

If the EA is set to trade five positions, all five long would have entered. However, after refreshed the time frame, only two positions appeared to have entered. So this looks much better than reality in history when you eye ball the chart.

Just for information if this is not known to you guys.

Hi K,

Thanks for pointing this out.
For me:
Your point about number of positions is a absolute critical consideration
in the overall framework of how these programs can produce some $$$$$$.
I have been doing trial and error with various configurations....
I got some heady ideas from Bill here...

http://stevehopwoodforex.com/phpBB3/vie ... &start=100

So I'm thinking about positive swap ..with trend...1.5 hour trading window..2 sessions
MPTM...independent profiles and deployed with extreme prejudice

[Geges]

I have a test for 7:59 London setup ....8 pair positive swap with trend global closure $150.00

[sim2505]

Hi all,

First big thanks for Bob and Steve for the sharing of their strategies

It would be better to enter only with on Trend signals and close on CT signal ? But it's maybe me but I don't see an option to do that. We don't need to trade all pairs, just those in trendy ?
A swap or CSS filter or human eyes could filter better pairs and only enter on Trend signals ?

Thanks for your help

:hi:

[Gertje]

Account size and Ball size.

I will stick to my 0.01 lot trades.

How did your trades do today? I also closed early for a few pips profit. And will try again when Asia opens.

It's just demo.

Goal is to gather info about best time of day/ day of week/ max DD and so on.
Then fire up 2 or 3 Empty4's with the same account and put differnt settings on each Empty4 that complement each other.
Could be different TF, pairs, settings, you name it.

On the other side, if a bot capable of making over 10% a day, comes with a max DD of 25%, would you discard it?

[SteveHopwood]

Just notice a phenomenon that when I change time frame from H1 to H4 then back to H1, some of the arrows disappeared upon which position has been entered based on the signal, particularly if you are trading multiple positions. Therefore when you eye ball the chart history, it looks much better than reality.

I was going to post about this today, because I have had time to think about it and can explain it.

The formula generates a signal and prints an arrow. The market conditions that generated the signal can change quickly, but the signal was generated and so the indi does not delete the arrow. If you then change chart time frame and back again, there is no signal and so no arrow.

Another drawback with the indi is the position of the arrows. These are nowhere near the market price and so appear more favourable in hindsight than they actually were - the distance between buy and sell arrows is greater than the trade entry and exit prices.

None of this negates the power of the indi; it just isn't quite so miraculous as it appears on our charts.

[romsyam]

Account size and Ball size.

I will stick to my 0.01 lot trades.

How did your trades do today? I also closed early for a few pips profit. And will try again when Asia opens.

It's just demo.

Goal is to gather info about best time of day/ day of week/ max DD and so on.
Then fire up 2 or 3 Empty4's with the same account and put differnt settings on each Empty4 that complement each other.
Could be different TF, pairs, settings, you name it.

On the other side, if a bot capable of making over 10% a day, comes with a max DD of 25%, would you discard it?

No chance. That's good enough for me. It's equal to 1% a day 2.5% DD. Looks tastier when you write it like that

[piet.forex]

Account size and Ball size.

I will stick to my 0.01 lot trades.

How did your trades do today? I also closed early for a few pips profit. And will try again when Asia opens.

It's just demo.

Goal is to gather info about best time of day/ day of week/ max DD and so on.
Then fire up 2 or 3 Empty4's with the same account and put differnt settings on each Empty4 that complement each other.
Could be different TF, pairs, settings, you name it.

On the other side, if a bot capable of making over 10% a day, comes with a max DD of 25%, would you discard it?

Gertje
Apologies if my comments sounded like criticism, it is not.

I am a very bad demo trader. Jumping in and out of strategies all the time.

I have much better discipline with my live accounts so I trade very conservative.

Keep on with your research. I learn a lot from you.

[SteveHopwood]

V 1g is in post 1.

I am trading the majors plus gj on the H1. I have seen groups of trades rocket to about 150 pips profit in a matter of minutes. The way the signals are generated will result in a lot of that profit being given back before SB closes the trade, so I have mptm set up to close the basket at $15.

Trouble is, this can result in an immediate re-entry at a far less favourable price if there is a signal on the current bar. 1g addresses this buy refusing a trade if one has already closed during the current candle.

Add this if you want to make the changes yourselves.
Copy this function into the body of the file:

Code: Select all

bool DidTradeCloseThisBar(string symbol)
{
   //I am using mptm to close trades as a basket and so do not want an immediate re-entry.
   //Returns true if it finds a closed trade on this bar, else false.
   
   int ot = OrdersHistoryTotal();
   if (ot == 0)
      return(false);
   
   for (int cc = ot - 1; cc >= 0; cc--)
   {
      if (!OrderSelect(cc, SELECT_BY_POS, MODE_HISTORY) ) continue;
      if (OrderSymbol() != symbol) continue;
      if (OrderMagicNumber() != MagicNumber) continue;
      
      if (OrderCloseTime() >= iTime(symbol, TradingTimeFrame, 0) )
         return(true);
   }//for (int cc = OrdersTotal() - 1; cc >= 0; cc--)
   
   //Got this far, so no trade closed on this bar
   return(false);

}//bool DidTradeCloseThisBar(string symbol)

Then go to bool IsTradingAllowed() and add this snippet:

Code: Select all

   //I am using mptm to close trades as a basket and so do not want an immediate re-entry,
   //so do not trade if a trade closed during the current candle
   if (DidTradeCloseThisBar(Symbol()) )
      return(false);