Simple trending

To add comments, please log in or register
Yuri Predborski
561
Yuri Predborski  

The Idea

We start our trending with entering a position. It can be whatever your heart desires, whenever and whichever way you want. There is one condition: its a start. After you started, you must open new positions as soon as old ones are closed.

The lot size is 0.1 for first order. TP and SL are 100 by default. You can use 50 and 50 or anything else as long as its not less than 40 and TP=SL. If you are going to use other values, you should know for sure they will work fine.

When we hit a TP, we set lot size to 0.1 and enter a new trade in the same direction immediately.

When we hit a SL, we double our lot size and reverse our position immediately.

We want to always be in the same way as trend so if the trend goes against us, we reverse.

That is the Idea.

The EA

There are 2 versions of EAs. 0.6 (0.3 with fixed coefficient, 0.5 with fixed bug) and 1.2 (fixed bug).

Both versions, 0.6 and 1.2 are valid till October 2007.

0.6 is used "as is" - you will need to set all parameters yourself.

1.2 contains all settings for all the pairs Alpari offers in forex group, consisting of 23 pairs. The data is written in so you can just load it on 23 charts and the EA will pick settings itself.

You can use 0.6 instead of 0.5 if you set the same magic number.

You can use 1.2 instead of 1.0 without changing anything.

Make sure you set different magic number for all currencies.

Time-frame independent.

Settings for v0.6:

TP.CT - take profit level.

SL.CT - stop-loss level.

IncreaseCoeff = 1 by default, each time we have loss increase lot size.

Coeff - increase lot by this coefficient.

Settings for v1.2:

UsePreSet - 1 by default, will set all settings depending on the pair. If the pair has not been found in the data base, it will use settings you put in. Set to 0 to always use your own settings.

If you get any setup questions, problems or bugs, feel free to ask here or pm me.

Demo Login details:

Alpari MT4 (you can find it on alpari.org)

http://alpari.org/files/mt4setup.exe - download link

Version 0.6

Login: 516307

Pass: 4eljazu

Settings are: 50 TP, 50 SL, IncreaseCoeff 1, Coeff 2, BasicLot 0.1, MagicNumber is from 1 to 23 depending on the currency.

Version 1.2

Login: 516254

Pass: z5nugbb

Settings are default, user's options are ignored, build-in settings are used instead.

Both accounts have been reset, both will now have $200 000.

There was a bug in the EA. It was fixed but needed a restart. I'm posting updated versions

Current situation

Current situation will be updated on every weekend since there isn't much interest in this EA. I'll just let it run in the background.

Demo Account v0.6.

Current max margin:

35 000+

Current balance:

211 000+

Current equity:

211 000+

Demo Account 1.2.

Busted. I'll make another MM for it (safer) and make the last try.

The problem was with lot increase on 2 pairs that got choppy and killed the deposit. One went to 79+ lots and the other to 50+. I'm making another MM where this will never happen. If that one gets busted, I'll only keep the first one in educational purposes.

I'll update with detailed margin and undetailed balance and equity (1000 step).

Features to be added:

Stop trading some currency if specific level of loss (number of stoplosses in a row) is reached, resume trading this currency after it has moved at least 50 pips in one direction (to avoid small ranges, feature taken from Nature System by Dan).

More to come.

New tool!

I wrote a simple program (EA) called MaxMargin2 (first version only wrote the highest margin it could remember, version 2 doesn't forget anything). You can have it for free. This tool monitors your current margin and writes it out in comments. It also writes down into a file the max margin you ever had (if you get a lower margin, max remains the same and when you get new max, file is updated). It won't become 0 when you restart MT4. If you feel like supporting my work with a donation, PM me Oh and don't worry, the first time it starts it needs to create the file so it will say "error 4103". Only once though. If it gives more errors, write me.

wolfe
1037
wolfe  

Sounds interesting, I'd like to see the results of that. This shouldn't be very complicated to program. Give it a try. 100 pips could be too long of a trend to keep up after your first TP is hit on some pairs. My experience with doubling after a loss is this: At some point in time you will not have enough margin to cover the huge lot sizes and draw downs that will occur. Doubling of lot sizes happens quicker than you might think. Just my input.

Michel
476
Michel  
Shinigami:
Hi everyone.

I have an idea of a system that might work and I'd like to share it.

The system is really simple. We place an order (any direction). We have 100 pips stoploss and 100 pips takeprofit. When we have a TP, we open a new order in the same direction. When we have a SL, we reverse. We think that if we had a stoploss, then we entered against the trend. If we had a TP, the trend will most likely continue.

I was wondering... What if we double the lot size each time we have a SL?

Once I heard of a similar strategy, it ran on 11 pairs for a year with initial deposit of 100 000 and starting lot size of 0.1 lot. The result great.

Anyone willing to try this strategy?

Are you able to make an EA or do you need some help ?

Yuri Predborski
561
Yuri Predborski  

Nope, I don't have an EA, but if anyone is interested, you can write it to see if it gives a good result. I got my hands full with few other EAs that require a lot of work.

jlpi
360
jlpi  

I did something similar in a form of a grid EA with intervals being 70 or 100 depending if it is trending or not.

Basically when it was trending the orders were opened only one way and when ranging both ways.

It is maybe not exactly what you want but you can use this EA as a start.

It is in https://www.mql5.com/en/forum/176535

nicesurf
369
nicesurf  

The problem is that if the market just move in a low range.....

Nic

Yuri Predborski
561
Yuri Predborski  
nicesurf:
The problem is that if the market just move in a low range..... Nic

If there is low range, you just set a high SL and TP and you'll sit and wait.

Yuri Predborski
561
Yuri Predborski  

Ok I finished the EA. I'll attach it to first post.

I also started a demo account. Balance is 1 000 000 and I put 12 EAs to trade 12 pairs. Alpari MT4. I'll put login-pass on the first post as well.

Wait a month or two and we'll see what we can see.

TP=SL=50

geez sorry forgot to use edit button....

bluehippo
31
bluehippo  

Hi Shinigami

Once I followed this thread here:

https://www.mql5.com/en/forum/177047

with an unfinished EA here:

https://www.mql5.com/en/forum/177047/page6

not very different from your idea - but also very different.

With lotsize increasing (fibo)

Its a martingale strategy but: it dies in small ranging not in outbreak like others. I tested it some time on GBPJPY but the EA tests very slow so I gave up

So what is easier?

To prevent an EA from getting stuck on the wrong side of the trend moving hundreds of pips into loss

OR

to prevent an EA from increasing lotsize in a small market range using some kind of indicator or whatever.

My feeling says that the above system (not very different from yours) can become a winner, have some ideas in my mind.

Yuri Predborski
561
Yuri Predborski  

Hi BluePippo, good to see you here. Thanks for the links, I'll take a look at it.

Oh and a funny thing - I just started and (as you would expect) I'm in big minus. But the balance already went up I'll have to check what SL/TP (the best combination seems to be SL=TP) is good for each pair out of those 23 alpari MT4 has (they say there are 28 but I only found 23). I found out 80/80 works good with GBP/USD but I don't have enough data to test all currencies. I would do it all myself but I only have data on EUR/USD in one of 7 running MT4's but I don't remember which one is it. I can download all the data for all 23 pairs and I'll do so later but that will take the whole weekend to install it (remove old data, put in new data, convert from M1 to M5, M15, M30, H1, H4, D1, run a MACD sample on EACH timeframe EACH pair, that will take the whole weekend for sure).

I'd love if you people could help me with that. I need all currencies tested with these sets: 30-30 40-40 50-50 60-60 70-70 80-80 90-90 100-100 110-110 120-120. If you can test it for last 3 years for few currencies, please do. Post your results here.

I think there is no need to test further than 120 because it takes too long to finish. Also there is no need to test small values - lot size becomes too big for my 1 000 000 testing account.

That was quite some work but I'm done. I now have enough data to pick somewhat good settings for each currency pair. I downloaded all the data Alpari has to offer (the quality is horrible, if you know a better free source with few years of data, please let me know). That would be enough to test all the currencies with different TP-SL and to see how well my EAs work with them. There is a big problem, however. Alpari now has only 2006 and further data. So I am only able to backtest last 1,5 years. Besides D1 data's modeling quality is 66-67% for majors and major crosses and 10% for new currencies (like NZD). 89% for H1 though (I don't use higher TFs). I'll post new login details for an updated demo account on monday. I'll also upload new version here.

If anyone still wants to help, help me find high quality M1 data for the last 7 years in free access for major currencies and their crosses.

Yuri Predborski
561
Yuri Predborski  

I finally managed to get new version up and running (I've been testing and picking settings for all pairs up till this moment from friday. Its been a tough weekend for me)

You'll find new login details in the first post.

You'll find new version in the first post later today.

The biggest improvement - I've coded all the settings into the EA.

You just attach the EA to one of pairs and it picks settings for itself by itself. Just don't change anything.

Or you can take it and run like 3 days of backtests (I generated 20 gigabytes of data while backtesting) and then you'll certainly get better data.

I'll be forward-testing this specific EA for the next 2 months (some pairs need a lot of time to generate any profit) and after 2 months I'll be able to tell if it really works. There may be few pairs that are good to trade and few that are bad to trade using this strategy.

Anyways don't expect any major updates during next 2 months.

And it will only work with brokers that use standard currency names like EURUSD and USDJPY (if there is m in the name, it won't work and you'll have to pick settings yourself!).

Don't go with this to a live account, its just a suggestion that is being tested.

Also if you are going to test this as well, use Alpari MT4. I cannot guarantee work with any other MT4.

To work with more than 1 currency pair, attach the EA to exact currency pair. The EA picks settings for the pair automatically (if you didn't change anything). The database contains settings for 23 pairs (excluding the very new ones) for Alpari MT4.

123
To add comments, please log in or register