the arrow indicates an inner bar. the alert goes off when the system gives a trade signal. the arrow is a prerequisite for the alert, but having an arrow doesn’t mean that there is a valid trade signal.
Darez it was great working on this with you today. I have some comments based on back testing I did to improve some things. If you are on later tonight (gtalk) pop me a message.
That would be a nice amount of cash to make daily if this thing can produce positive results throughout the day, but i think because of the market slowing down for the holidays as well as it being off-hours and not during busy time that could reduce the amount of alerts.
Just warning you guys not to optimize it too much, as “curve fitting” is usually not the best idea when backtesting systems.
Very interested to see the finished product though. I downloaded it the other day, made some tweaks, and found that it worked better on longer timeframes. I’ll be able to colaborate more this weekend.
With all the knowledge I’ve gained so far about indicators and EAs the past few weeks, I’d like to try to make an EA for the cowabunga system again. I failed miserably before. The indicator on this forum doesn’t always seem right either, so I think I’m going to stay away from that.
yea thats generally what I do. I let it open trades automatically, but then I watch the market. And once I think it has gathered enough (or lost enough if its going in the wrong direction), I exit. When exiting manually I also take into account some of the other indicators, although James’ system doesn’t use anything else.
The stop/loss and take/profit numbers are something i’ve made up. The system does close open orders automatically when it sees a signal in the direction opposite to the current open order.
It wasn’t curve fitting, Initially it was making invalid entry and exit points. I think Darez did a great job on it so far but I think it could use some more work to fine tune it.
It still seems to be missing a lot of valid entry points and exit points in both back testing and my live trading, even with double checking everything is right.
I’d like to see the addition of the outerbars as well as well as the other reversal point (candle stick pattern) although as I said earlier to Darez it seems like it would be very tricky to code.
Possibly the ability to use multiple alert indications ie different graphics and sounds depending on what’s happening at a given time.
Open sound and close sound would be a nice touch.
Darez if you get a few minutes and if you feel like chatting let me know I wanted to go over a few things I found while testing as well as some additional things I touched on here.
It does actually leave the trade if a reverse IB forms as per the system. The tp was placed in there to lock in profits on a large move with no reversals to close it out.
Read the original thread about the system as it contains a wealth of information regarding stops, reversal points etc. The EA is great but you should know the fundamentals beneath why and how it works too.
Looks like today the system is doing everything backwards.
Many alerts most of which were losses. I am looking at the trend lines as well as some other indicators because I couldnt believe the thing wanted me to buy when it was going down USD/CHF. Hmmm… did you get this too?
I am thinking that keeping with the positions open is a good thing because they do change, it takes a couple hours, but i was able to get most of the losses back in the demo account so it does work, regardless of how long it may take. even profit was made ultimately. Setting a good stops and T/P and walking away will probably end up working if using this with other indicators etc.
tonicfan, you cannot blindly follow a system… any system. You have to use your own knowledge as well. If there was such a thing as a perfect system everyone would be on an automated trading facility. And yes, the IB system does fail… through back testing i’ve seen places where it indicates a trade in one direction whereas it clearly looks like a trend in the opposite direction. This EA simply implements James’ system, it does not claim the systems effectiveness. Our job (and Jimmy and I are working on it) is to look for other indications and improve upon the system.
Now, having said that, i did not get any buy signals for usd/chf. The only signal i got for that pair was a sell. Here’s a screen shot of today’s trades using the system, with the usd/chf pair trade high lighted.
Welp I got a buy signal, so i guess its not the same for you.
Im not blindly following this I am testing it and i am looking at my own trades for real and seeing that it is a little off, and was asking if you had a similar situation. Next time ill keep it to myself, sorry i was trying to be helpful.
its probably the time duration charts you are considering are different from mine.
Also, of course any help is appreciated. But the comments should be in specific terms like “at so and so time, such a pattern occurred which indicated such and such, but the system opened a trade in this direction”.
Saying things like “its doing everything backward” doesn’t really help
The reason it could be different and I notice this with a lot of demo platforms I test, is that some dealers may be using different data feeds or have delays etc. I normally have a few demo accounts running at a time checking things and notice slight differences between the candles. As an example… in the case of this system… sometimes an IB is perfectly formed on one demo account but not another.
Generally they are pretty similar.
I also think you should stick to more active (read liquid) pairs with this system as the spreads are usually tighter.
Are you trading off 15 min charts? Did you read all the posts about the system this EA is attempting to mimic? There is a wealth of information in those threads that will help you under stand how to better trade this system.
I also agree with Darez, auto trading any system is not a good idea.
I like this ea because I don’t have the patience to look at all my charts every 15 minutes. The EA watches for the signals and allows me to work on other things in the meantime. Once a signal pops up I can take a look and based on my experience and what’s going on generally in the news etc. I will make a decision to go in.
Thanks again to Darez for taking time to code it and James for posting the system it is based on.