I have the ae up and running but not long enough to take a trade. However, I did do a back test a few months and the ae started out great but came back and wiped out a 10k balance. An earlyier post stated to get rid of trade 2 and 3 and it will be profitable. I here the manual trading is doing well. Will try and see what happens.
The Bat...a strategy by Team Aphid
- #744
- Jan 17, 2008 5:07am Jan 17, 2008 5:07am
Hi All,
I just found this thread and let me say, its awesome!! Big thanks go to Team Aphid for developping AND SHARING this strat for free with us. While reading ant backtesting this strategy, it remind me of another good strategy which is also very good BUT hard to manage, as you need to be constantly monitoring the trades. Of course am talking about the ICWR Trading System.
So.. Both are very good strat but you need to be very disciplined and not try to bend the rules ELSE it won't work as expected.
All the best and THUMbS UP.
http://icwr.ovenstone.net
KaMpeR
- #758
- Jul 12, 2010 7:09pm Jul 12, 2010 7:09pm
Hey Jig,
Agreed, I think better not use MT4 though jforex dukascopy's programable platform would be better . there is a good tester with proper tick data too.I would be interested to form up a team and have been meaning to learn programming....how hard can it be.
Had my interest spiked by automated trading recently. we could hire someone in also.....
I liked the bat from the start v elegant system.
Si
Yeah my interest with automated trading has been spiked lately also, I mean if I spend 14/16hours watching the markets a day I don't really need to be staying awake during Asian session just to trail goddamn stops when a simple EA that would "close y above/below line x" would suffice !
. There's a load of other simple things I would love to have automated just to make my trading day easier. Aside from that I also believe there are simple mechanical edges to be obtained in the forex market and other markets, most of which I believe will be in the market as long as they exist due to microstructure and the inherent nature of the markets as generated by the participants.
I've also been meaning to learn programming for these reasons, I mean I spend enough time around the PC monitoring trades anyway so I could use some of that time to learn programming instead of wasting time. And as you rightly said: How hard can it be.
I concur that we should form a team, I reckon I would rather learn programming myself than hire somebody else as I like being self sufficient and "understand" what is going on, also I reckon it could be an interesting hobby :nerd: That being said I wouldn't rule out paying a programmer, as I am currently working with a programmer to automate another system of mine anyway
.I will look into the Dukascopy JForex as I am unfamiliar with it, I also agree with you about the simple nature of the BAT.
Best regards,
Jig
Also check your PM's
EDIT - You don't have PM's enabled, can you enable them/contact me. Thanks
- #759
- Edited 7:52pm Jul 12, 2010 7:27pm | Edited 7:52pm
It's an excellent system, as is, in my opinion. I wouldn't clutter it up with filters.
I traded it for about a month, and took every trade.
Trading it manually will run you nuts. You have to watch it almost every minute of the day. Pending orders have to be continually revised and costly mistakes are inevitable.
There was/is an EA for this system, that didn't work (no surprise there).
When I say filter I mean a simple automatic trend defining indicator, a simple 21ema 50sma cross from a higher timeframe would do the trick in my humble opinion. I don't want to get into optimising etc but something trend defining from a higher timeframe would improve expectancy.
2ND EDIT : Reading this back this seems rather arrogant of me which I do not mean to be, from reading Team Aphids posts they obviously backtest everything thoroughly, know what they are talking about and have probably tried this idea already, but from my initial (albeit relatively small in timescale) backtests a filter improves expectancy. I will manually backtest more tomorrow.
Although I see from the earlier posts in the thread especially the post/posts made by hanover and Team Aphid that they tried to filter out trades but by doing so what happens is you indeed do filter some losses but you also filter out big winners. My initial thoughts were that a filter would help immensely however I would like to see what the results would be on years on the 1HR on years like 2003 and 2006.
I saw the thread where the EA was going to be programmed and where no programming was achieved, it was strange to me that no success was achieved considering there's usually people programming all sorts of terrible ideas around this sub forum
.Just a few questions if you don't mind, What do you think of the number 2 and 3 entries as outlined originally ? Did you use them ?
I assume you stuck to the 1HR timeframe ? Thanks in advance.
Regards,
Jig.
EDIT - I agree with the "run you nuts" of trading something like this manually, It's like my trailing stops, it can't be that hard for me to help myself program a two bar stop ala the Turtles and stop having to look at charts and move stops every 15Mins.
I actually use the BAT for trailing stops on various systems as well, except I use a close above/below instead of the "touch"