MultiBob, a multi symbol NB10.2(a) autotrader

[Jollyroger]

Well been testing for a while now and the bot has lost about 20% of original balance. There seem to be too many losses compared with wins and the wins are not compensating for the losses. How are other people getting on lately?

I noticed that I had the Take Profits set at 100. I think this was the limiting factor - as this is a trend trading system, to make ends meet, we need to let winners run. I have now set take profit to 1000 and I will trail stop losses manually under the last swing high/low on the daily charts to see if that increases profitability!(unless there is a way of automating this that I have missed??)

I'm using all the option (10.2a, CTA etc). It started with promise. Maybe the sample pool is too small and it needs greater time over varying market conditions. These markets are tough to trade - even the large Forex Profits are down at the major investment banks this year the FT was reporting today. It said many traders were sat on the sidelines due to Central Banks' intervention messing up normal market conditions and trends.

I've given up on this bot and am concentrating on Bob's 10.3 scalper. I lost about 20 % also. I hung around for awhile hoping for a reversal but lost patience and closed multiple trades. Most of my trading was with 1.1.11.
Roger

[AGT]

Hi Jollyroger - the same results for me here ... I have given up too ....

[grimweasel]

The thing is systems have drawdown periods - I still think the essence of the system has merit. I just think that tweaking is required to ensure losses are killed quickly and profits are allowed to run. The Forex markets are difficult for most systems at the moment. I think I will continue to run the bot over longer periods to give it a chance?

[yamez]

MultiBob v 1.2.14 10.2a demo results from 22/10/12 to date, with no intervention. Broker is FX Primus ECN demo.

Reversals/major re-tracements hit it hard last week which will happen with a trend following EA. Set file in post http://www.stevehopwoodforex.com/phpBB3 ... 586#p28586

[SivEd]

Last few days have been much improved for this bot, anyone else seen this?

I am running three instances of 1.2.14 on one account (Finfx demo) with 10.2, 10.2a and CTS set @ fixed 0.01 lots. OOTB otherwise. Made 600 pips this week, 780 since Tuesday.

Previously loosing just like everyone else.

Begs the question can it be filtered to trade only when conditions are right? I agree with Grimweasel, this robot deserves a future, either with optimisation, and/or with some trade management to hold on to the winners and exit the loosing trades more quickly.

Cheers,
Pete

[yamez]

Results for MultiBob 10.2a, just over a month.

I am going to start taking automatic screenshots of the trades. Some entries do not break the daily pivot and are 30+ pips away. Most trades do enter on breaks, I just need to get a better idea of how many.

I have been manually using CTS (http://www.stevehopwoodforex.com/phpBB3 ... f=12&t=875) for filtering the signals from Multibob 10.2a. Since I'm still getting my head around CTS, I'm just focusing on trend trades.

Looking at some past charts using 10.3 DO (http://www.stevehopwoodforex.com/phpBB3 ... f=38&t=919), NOT taking 10.2a trades opposite to the DO would prevent some trades being "caught" in a retracement/reversal.

Such a filter would be easy to program into the EA but us testers should see if any filter has merit before asking Baluda to code.

[effluvium]

The thing is systems have drawdown periods - I still think the essence of the system has merit. I just think that tweaking is required to ensure losses are killed quickly and profits are allowed to run. The Forex markets are difficult for most systems at the moment....

I absolutely agree with you Grimweasel. Multibob EA needs some tweaking to avoid a huge DD.

I don't know why the lot size I put in this EA is not taken as a maximum size for a trade and I see some lot size 3 or 4 times bigger than what I specified

[yamez]

Latest results. All this is quite meaningless without some walk forward testing if that were possible. My demo account is also overrun with EAs so that is affecting the margin limit filter too.

[kwanann]

The thing is systems have drawdown periods - I still think the essence of the system has merit. I just think that tweaking is required to ensure losses are killed quickly and profits are allowed to run. The Forex markets are difficult for most systems at the moment....

I absolutely agree with you Grimweasel. Multibob EA needs some tweaking to avoid a huge DD.

have been running MB on live and demo for about a month on fxprimus, probably not 100 accurate, but i do realize that there is a kind of DD trend. DD tends to go up in the early market opening hours, hovers around max dd about 8 hrs from market opening and goes back up in the later part of the day. Currently i'm trying out manual trading following these trading hour rules and see how it goes

I don't know why the lot size I put in this EA is not taken as a maximum size for a trade and I see some lot size 3 or 4 times bigger than what I specified

I think you forgot to turn off the risk %, if risk % > 0, it will take lots depending on the risk %, if you want fixed lots, set it to 0

[Baluda]

I can not resist putting this up.

MultiBob.png

[/url]
It seems MultiBob takes off when the market is trending. This is MultiBob v1.2.15 OOTB (unreleased but essentially the same as v1.2.14).

Paul