"better"-EA Automated trading championship REVEALED - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Webinar today at 12.00gmt in tradingrooms.com
- community rooms
-MaTRiXZoRRo room
iGoRHello iGor,
thank you for your webinar, it has been an eyeopener! Wish you would continue this great work.
It is a MUST for every (wannabe)trader.
best wishes.... Dirk
For those who would have missed the webinar here is a link to the recording:
http://www.mediafire.com/file/tj5loqeuywm/PublicWebinar_EA&General_Info.exe
My special thanks goes to Arctic_4x who made the recording.
The recording is a video file. It is a webinar where one can see what happend LIVE in the video room.
regards...iGoR
sorry forgot to post the spreadsheet with the results of the "Better"-EA.
This is the spreadsheet that is mensioned the webinar.
regards...iGoR
I can't help but be a skeptic on this whole thing. So the guy made some really good money (fake money) over a 3 month period. The concept is there, I agree. But we are sampling 3 months of data. I have EA's that have made more than that in 1 month. The next month, went broke. When you can give me 2 years or more of data, live forward testing, I might believe it. Like I said, the idea was good, but realistically.... you can't predict the future, only base ideas on the past.
I can't help but be a skeptic on this whole thing. So the guy made some really good money (fake money) over a 3 month period. The concept is there, I agree. But we are sampling 3 months of data. I have EA's that have made more than that in 1 month. The next month, went broke. When you can give me 2 years or more of data, live forward testing, I might believe it. Like I said, the idea was good, but realistically.... you can't predict the future, only base ideas on the past.
As I explained in the webinar and gave proof of, even after 10 years it can go the wrong way.
There will never be an EA that will work properly till infinity....
regards..iGoR
As I explained in the webinar and gave proof of, even after 10 years it can go the wrong way.
There will never be an EA that will work properly till infinity....
regards..iGoRI saw the webinar , your point is that Better Expert was optimized on a bullish market, that as we know is a problem that can be solved when training the expert on an alternative bull/bear data , then test it on another data.... but the man is losing money now so I think he is working hard to figure out the problem, maybe he should have seen your webinar.
while its very difficult to make an expert that works on all market conditions, but we can make an expert that can switch between the different market conditions , In engineering we learn never to say never. It start with a dream and desire, the rest is hard work, but I agree its a rather difficult and we may say its almost impossble to make such an expert.
MT4 is a small toy, central banks are trading forex and invest millions of dollars for smart trading algorithms, so IFthey can have figure it out, I guess one day we will.
I saw the webinar , your point is that Better Expert was optimized on a bullish market, that as we know is a problem that can be solved when training the expert on an alternative bull/bear data , then test it on another data.... but the man is losing money now so I think he is working hard to figure out the problem, maybe he should have seen your webinar.
while its very difficult to make an expert that works on all market conditions, but we can make an expert that can switch between the different market conditions , In engineering we learn never to say never. It start with a dream and desire, the rest is hard work, but I agree its a rather difficult and we may say its almost impossble to make such an expert.
MT4 is a small toy, central banks are trading forex and invest millions of dollars for smart trading algorithms, so IFthey can have figure it out, I guess one day we will.Hi Minime,
A system that I showed in the webinar worked well over a 10 year period as well in bullish markets as in bearish markets as in longer periods of consolidation and still after 10 years suddenly it finshed working properly or showed drawdowns that were bigger and longer then any drawdown over the previous 10 years.
The EA that would work properly all the time needs to identify every sort of market change. Not only in bullish or bearish or consolidation but the frequenty of the moves or the dynamics itself.
As I said in the webinar if that is indeed possible that an EA is possible of doing this then this EA needs to work on every single pair index commodity future etc....
I showed clearly on the day charts that the cable and jpy at some point started to behave completely different.
The moment someone says I have a very good EA that works well on the euro but does not work properly on the JPY then you can forget about it because for all the same the euro can start to move over a certain period with the same frequenty and dynamics as the jpy. That means that over that period the EA will loos money. If that period is only a couple of days or weeks no problem. If that period is months or years then the account is burned.
It is against all logic and common sence that we see in the normall world that this would be possible.
Take a car...
If one want to go to the race track you use a F1 car. If you want to go into the mountains you will use a 4x4. If you would go into the dunes or the beach you would use a sand buggy. If you have a big family you will use a station wagon. If you need to carry a heavy load you will use lorry.
The moment that you would be on the race track and they put a sand dune in the middle of the track and you need to take 5 people on board you can forget it...
Everything in life needs special tools or instruments to work to its fullest capacity. The moment the conditions or enviorment start to change you can forget it.
And market conditions have the tendency to change suddenly. If you have a specialized tool or EA that can work to its fullest capacity on particular conditions or specific item, it will get in trouble if the enviorment or the conditions suddenly change.
On day we will ?...as you say it want be with a MT4 platform nor with omega tradestation nor amibroker nor metastock professional. It will require at least a big blue pc from IBM to hold the software that can turn a system that can recognize all the changes in the market conditions and then the same problem will occure again that this recognition will have lag and that it will recognize it after the facts.
The only solution to this (IMO) is that one learns how to trade manually, discretionary and train the eye to recognize the changings in market conditions and be prepared to addapt yourself. That addaptation is possible if one has enough knowledge where he can depend on.
That knowledge comes true a lot of trading hard work and lots of time....and not by looking and surfing the internet and forworth testing every possible EA out there ( as many people do).
regards...iGoR
Hi iGoR,
You have some valid points , but from an engineering point of view ... infinity is a number and impossible is something that is waiting for an engineer to figure it out.
It is true that EURUSD can behave like the jap pairs but there are certain things in forex that even if they change they will come back to nature , for example the correlation between pairs , the daily range for some pairs , the spread on the EURUSD ... those are keys that can be used to overcome some problems if an expert went in troubles .
anyway thanks for the presenation , it was useful
QUOTE=MiniMe;231880]I saw the webinar , your point is that Better Expert was optimized on a bullish marketl....
I listened to Igors webinar on monday, and he made some very interesting points, particularly his observations regarding the possible correlation between the systems equity curve, and the exchange rate of the euro. It was definately worth taking the time to listen.
However I dont really agree with this "optimised on a bullish market" theory. Its been a while since I looked at betters trades, and I really should take a look at Igor's latest spreadsheet before commenting, but the last time I looked, the number of long v short trades was pretty much 50/50.
Longs achieved 134 winners out of a total of 205 long trades, a 65% strike rate, whilst shorts achieved 125 winners out of 203 short trades, a 61% strike rate.
The average size of winners for longs was around 38 pips, whilst the average price of shorts, around 24 pips resulting in a total of 2446 pips for long trades v 1942 pips from short trades. Whilst this looks significant, the distribution in sizes of gains and losses from both long and short are pretty similar, and the differences can be attributed to a few outliers in each sample.
The argument that EURUSD was in a strong trend on the weekly chart just does not cut it for me. We are analysisng trades from an EA whose stops and targets are typical of a one or five minute timeframe. Over the course of the competition, the EA exhibited a wide cross section of market behaviour. It performed well in upwardly trending markets, it performed equally well in downward trending markets, I see very little evidence of bias, other than the EA was able to capitalise on a couple of strong up moves, if the market had experienced strong moves south, theres plenty of evidence to suggest that EA was perfectly capable of capturing these moves too.
Having discounted Igors theory, here's my take on whats happening with this EA. If I remember correctly the EA uses 3 different sub systems, and the trades within each sub system alternate in a strict pattern of Long, Short, Long, Short,Long, Short etc, so if a sub system is currently long, the next trade taken by that sub system will be a short even if a long setup and trigger presents itself prior to the next short set up. That's a pretty strange approach which suggests all manner of interesting strategies are in play. This behaviour isnt immediately apparent until you filter the results by sub system, then it becomes clear.
I have a suspicison that this approach possibly makes the EA more susceptible to getting locked in or out of phase with price action, when it locks in phase, it does great, when its out of phase it does not do at all well. Most TA based systems suffer from similar phase issues, Im not sure if imposing a strict order makes the situation worse or better, I can think of arguments for both cases.
Its a real pity that better wasnt there to defend his case, it would have made a very interesting webinar.
...
Attached a little document dating from 1994 (if I am not mistaken 486 were the strongest PC's then)
Could post some more similar things, but I think there is no need
To clarify : I am not a NN skeptic, but seeing the magic the simple NN mentioning does to people makes me wonder. Right now it is more like a religion than serious investigation. This document is just a simple voice from someone that really did investigate and worked on NN. Might be worth reading (no math, no complicated explanations, no super-duper high end, clever only stuff).