Another $100 Experiment

Hi,

I'm BrokeTrader and this is my EA journal.

A part from the fact that I am, by far, a better programmer than a trader, I have a full time job and (un)fortunately so, I have no time to trade live.

That's what motivated me to learn MQL and to spend great effort and time in developing this EA.

As a trader, I went through all the phases I think any trader has to go. The luck of the starter, the greed of the dreamer, the fear of the loser and the desperation when seeing one account blown after another. These years were sometimes funny and sometimes depressing but at the end I learned one thing about myself and it's this: I am not reliable as a trader, period. So better spend my time to write some code to trade for me and eventually enjoy the performance if any. And that's my current status.

After several failed tentatives in trying to write a profitable EA, I finally came with this latest one which is as of today my best piece of code in automated trading. It still has to prove that it's durable and profitable.

The EA is a mix of initially three algorithms which, after post 42, were then expanded to eight and again later were expanded to twelve !
Twelve different algorithms generating twelve different and often contradictory signals to trade. Each signal algorithm being then connected to a separate or common trading algorithm that handles the signals to enter and exit trades according to specific rules.

If you want to know a bit more about my programming techniques and on how is this doable, don't hesitate to have a look at this thread.

As for the money/recovery management, I'm exploring a special approach of the martingale logic which I think has some edge, but I agree, the martingale thing is a wild horse difficult to tame. I think I did it to a certain degree, but only time will tell.

I consider this as an ongoing live test. And so, a considerable amount of time, several months and years, will be necessary to assert the success or failure of this venture. For this, I have created a special calendar as shown below.



Each box of the second line represents a month of trading. I have arbitrarily specified different stages that go from luck to safety. At the same time, more than at least one thousand of trades are also necessary to validate each stage but that should not be a problem given the number of signals that the EA provides on a daily basis.

All the steps and stages will be recorded here for further post-mortem study, either in case of failure or success.

BrokeTrader.

First crash !

When I though that there was a bug in my EA, I stopped it just to discover later that in fact there was no bug at all, unfortunately I derailed everything and couldn't recover and it looks like without my intervention the EA would have done well. My bad




New start. This time, I will let it run without touching it...
The EA is calibrated to earn 5% per week so this is my new weekly target.

Great man...
I think in this case..you should stay in front of your computer all the time?
so you will not miss anything from your EA

I don't understand your statement or you are being sarcastic ?

I just controlled each of the trades taken and there were two that I didn't understood and thought they were from a bug, that's all. Only at a later time I saw that they were perfectly justified.

I Never tested this EA live, only on backtests, that's why I was only double-checking, not being in front of the computer all the time.

nevermind sir..pardon my bad english. didnt mean being sarcastic at all.i said great because i started with $100 too...

No problem Gentlemen.

English is not my mother tongue too so I could also misunderstand you.

Wish you all the success with your $100 adventure, I will follow so don't hesitate to talk...
Maybe we could soon open a club of the $100 starters

It seems that I still have to learn to not to be greedy trading over my EA which does just well for the time being.

It was Friday evening and I just wanted to add some high leveraged pips but the only thing I did, in fact, was just risking to blow the account.

Sometimes I ask myself if I just look for self sabotage and why ???



Fortunately the luck was on my side this time and I could finish the week with green stats.



Terminated the week with 22.4% profit which is of course good and over my target of 5% but this is due to the fact that the account is under capitalized so this will normalize as the balance grows.

My back tests show that the the minimum balance should be something around $750 so until then the risk will be greater than normal and I'm wondering if I should not capitalize my account correctly. May be I will do it when I see that the account is in jeopardy.

It was looking like I had to refund my account or let it implode and start again but it seems that it was not yet the time and the EA recovered at the 6th tentative. This time I didn’t put my hands on it, lesson learned last week.

Checked 38%+ drawdown in line with what backtests show (50%+) so nothing abnormal here given the tiny balance.

And here I am, with green stats for this week and prepared for next week challenge.

Next week I will trade with an improved version of the EA on which I have worked all this week. I could increase the profits dramatically while keeping the same drawdown and risk.

To be continued…

Hi,

Would u mind sharing the entry and exit strategy?

Generally, the strategy

Thanks

Hello, Finally I opened myself a trading jurnal too..and like yourself it started $100.
Go go $100 Club

Tentative #2 failed, RIP



Didn't have even the time to put new EA to test.
Observed that I kept a cycle running through the weekend when it should maybe have been preferable to not start a new cycle on Friday. To be tested...

Next tentative starts tomorrow at 0H00 with new EA

sorry to see this mate..but i see u keep the spirit on.
thats the best way..dont worry everybody

Thanks azmoet and don't worry, it just means that I still have work on the table.

I knew that the account could blow for different reasons that I also knew.
As I said, I can live with that, so no problem so far. I will prepare and be ready for the next launch, it's not time to give up yet.

I just saw that your explorer is green as you are doing a good job. Congratulations, continue like that mate !

Hello broketrader, sorry to see that your second attempt failed. I'm also a fan of Martingale strategy, and I have developed an EA that trades intraday breakouts with maximum 3 Marti levels. I'm going to open a trading journal for it soon. However, we all know that Martingale strategy works only in two cases:
1) with unlimited deposit.
2) with an edge that makes sure there will be no more than N consecutive losses, which means if N consecutive losses happen, then, for sure, N+1 trial will be profitable. But this is equivalent with a Holy Grail system: if we know that after N consecutive losses, the next trade would be profitable, we don't need to do martingale at the first place! We just need to sell all our properties, and wait for the N consecutive losses, and then invest all our money into the next trade. Unfortunately, there is no way to guarantee that the next trade will be profitable. So, soon or late any Martingale strategy with limited deposit is doomed to fail.

Nonetheless, good luck with your Martingale attempt!

Thanks Math !

When you open your journal, for sure I will follow.

I'm done for the week.

50% in a week and 42% for the month with only one week traded ! What more could I ask, so far the best performance and it even survived the FOMC !

This makes the month of May an interesting one if the EA doesn't implodes in between...

Looks promising and moreover I have already found niches for improvement that will be at play starting next week !

nicely done mate..!!
if you can repeat this for a year..then its a gold mine.
best wishes for us both

Yes, if only...
But for the time being, the record is 15 days ;-)

The terminated month of April gave me one additional month of recent data for my backtests.

By running the backtests during that month, I could verify that the EA crashed the same way, and twice, as it did on the real account.

My finding was that I have, if possible, to terminate the cycle before Friday and avoid some trading on Fridays as it seems that volatility and behavior is quite different on that day and it's anyway a good thing to avoid open positions during the weekend.

All this has been programmed in a new version of the EA which is now in version 1.02 and which passes successfully the Month of April.

That's not to say that there will not be future black swans, I will just take care of them, the time I discover them.

Good thing that the 2nd week ended with account nearly doubled, this shows that the EA has some potential but that's only its second step in its infancy.

Moreover, I have updated it and corrected some minor bugs, it's now in version 1.04 and in the process, the Draw-down and profit have been improved but it still trades at maximum risk.

Now it's time for me to have some holidays, it has been two intense weeks of work, professionally and also on this robot, I need sleep time now and to cool down my head. Traveling through Italy for a week and a half will help but no holidays for the robot, that's the good thing with EAs, they are made to run 24/7 regardless you are there or not.

I just hope that the account is still alive when I'm back, fingers crossed.

Have good time and trade well.