MT4 & MT5 backtest - page 2

[Sergey Golubev]

This is CAC 40 Index, MT5, hedge account 9I did not use hedge but anyway) - 2 trades were closed in profit, and it is not backtesting - it is trading:

[Sergey Golubev]

Deevog:
Why your not accepting FIFO rules for Americans and MQL5 made this netting just for USA peoples and now brokers forcing all peoples to use this netting when they mention only netting accounts available. Its a worst mistake of MQL5. Concept is same from MT4 to MT5 upgrading few graphics and turning into 64 bit only new in MT5 is just netting to limited peoples whatever marketing MQL5 do. This netting and hedging should not be in the hands of Brokers.

The another question is MT5 can give more benefits to Traders that's why 90% brokers not willing to start this platform ? To change traders mind from hedging to netting wouldn't be easy possible it will take few years more.

Besides, I do not know why you want to compare backtesting results of MT4 with MT5 for example ... it should not be same ... and backtesting on MT4 with differet brokers (one strategy, one period/timeframe) - not same too.

So, you have a choice:

• to continue using MT4 as long time as the brokers will support it, or
• start using MT4 for example.

As to me so I am still considering MT3 with Pentium I and Windows '95 as the best combination ... but I am not using Windows '95 anymore ... and no any broker is supporting MT3 ...
It may be same situation with MT4 for example.
:) 

[Sergey Golubev]

Deevog:
I'm testing on Everytick only, I'm not using EAjust just to backtest,  Putting real money when over 5K and want to go more deep at least I can get 70% similar results as MQL5 shouting a lot MT5 more better then MT4. Second the major problem is with broker's I'm already working from last 3 years they have no plan to launch MT5 even MQL5 pushing a lot peoples to ask their brokers but Traders and broker both not taking interest , MQL5 all advertising going waste so far.

Everytick with real history not possible as mostly brokers not offering MT5. So depending just on EVERYTICK. MQL5 set netting function to follow FIFO rule just for USA peoples and few brokers set this rule for whole world as they not allowing hedging it's difficult to understand as MQL5 made peoples addict in hedging by MT4. Worst mistake from MQL5.

here you are
:) 
All (not yet) about Strategy Tester of Metatrader 5

Forum on trading, automated trading systems and testing trading strategies

All (not yet) about Strategy Tester, Optimization and Cloud

Alain Verleyen, 2013.08.20 19:47

A must ! Before using it or asking any question : please read the Online Help  (Also accessible by F1 on your MT5 platform)

Strategy Tester

For Traders

• General overview of the possibilities.
  

• Overview of results : Visualize a Strategy in the MetaTrader 5 Tester

• If you are thinking to buy a product on the Market : How to Test a Tading Robot before buying.

• If you don't see all strategy tester fields, they are probably hidden.

• How to use settings resulting from EA Optimization in your EA (without manually changing it). 

• New Testing trading strategies on real ticks.
  
  

---

For Programmers

• Detailed description of it works : Testing trading strategies (documentation).

• As noted by erdah, SendNotification, SendMail, SendFTP don't work in Strategy Tester (of course).
• IndicatorParameters() doesn't work in StrategyTester.
  
• OnChartEvent() in Strategy Tester.
• CopyTicks() function doesn't work on Strategy Tester.
  

• General description of how to use it : The Fundamentals of Testing in MetaTrader 5

• How ticks are generated : The Algorithm of Ticks’ Generation

• Discussion about Strategy Tester mode (1 minute OHLC vs every tick)

• How to test a strategy for all timeframe and compare results ?

• Using the TesterWithdrawal() Function for Modeling the Withdrawals of Profit

• How to use file with Strategy tester.

Optimization

• Description of the algorithm used for fast optimization : Genetic Algorithm - It's easy.

• Creating Custom Criteria of Optimization of Expert Advisors

• Running optimization from command line : https://www.mql5.com/en/forum/13468

• Don't expect too much from optimization and try to prevent curve over fit.
  

Cloud

Announce of the launch : MQL5 Cloud Network launched.

As Users

• How it works, an overview : Speed Up Calculations with the MQL5 Cloud Network

• Optimization by example : MQL5 Cloud Network : Are you Still Calculating ?

Some useful topic about usage of the Cloud :

• Error message : endless loop detected in OnInit function
• How much it will cost !

As Providers (agents)

• Frequently Asked Question about the Cloud Newtwork (FAQ) and How to Participate.

• When you start to provide agents on the Cloud, you have to be patient and use the search engine.

• An example of what you can expect as agents provider or 1 year of Cloud's agents providing 

• User experience.
  
  

 

Work in progress, stay tuned !

[Dua Yong Rew]

https://www.mql5.com/en/forum/23930

i can only share real trades vs back test trades 

[Deevog]

Sergey Golubev:

Besides, I do not know why you want to compare backtesting results of MT4 with MT5 for example ... it should not be same ... and backtesting on MT4 with differet brokers (one strategy, one period/timeframe) - not same too.

So, you have a choice:

• to continue using MT4 as long time as the brokers will support it, or
• start using MT4 for example.

As to me so I am still considering MT3 with Pentium I and Windows '95 as the best combination ... but I am not using Windows '95 anymore ... and no any broker is supporting MT3 ...
It may be same situation with MT4 for example.
:) 

Think When I need to start  30K USD account with a new broker who offers MT5 but before going to start I must need to test deeply what will happen, When you said 100 Brokers offers MT5  it's not easy to trust on mostly brokers they have a an Island regulations and only 10K trades required to be regulated there  is this quality. 

I depends only on FCA and ASIC regulated brokers and you know it few quality brokers not offering MT5 and I worked with them via MT4  and from last 2 years My accounts gained enough for me. All knows few brokers who offering Bonus and big leverage this is not trading quality.

Just saying 100 Brokers offering MT5  not makes sense, I think there an important decision before start MT5 or MT4  that decision is  are we secure with a broker with whom we going to invest capital. This all makes difficult how check results vis MT5 backtest , So if MT4 and MT5  with same strategy giving different results so on which result I can depend more? 

[Sergey Golubev]

Selecting the broker is very private thing: some traders prefer regulations, some others - location of the broker's office to be on same country or city, and so on. It is also depends on the fund the traders are going to deposit with them.

--------------

As to backtesting so I used backtesting on MT4 just to understand the following: EA works or not (by openning and closing the trades). If I wanted to have some idea about profit from the EA so I am trading it on demo with my broker for at least 3 months. But it is my choice: I believe in trading, not backtesting (in case of metatrader 4 for example).

If someone told me: "my EA is profitable because I backtested it with good results" so my question will be the following: "which broker?". Because everry broker is having different quotes/data (sslightly different). Example: well-known iTrend indicator is having differemnt value for different brokers, next example: famous Brainwashing EA - different results for different brokers.

That is why I like MT5: I do not need to trade by EA on demo for 3 months just to understand: EA may be profitable or not for that past period of time. I just need to backtest EA on MT5 with 'every tick based on real tick' mode with selected broker.

[Deevog]

Sergey Golubev:
Selecting the broker is very private thing: some traders prefer regulations, some others - location of the broker's office to be on same country or city, and so on. It is also depends on the fund the traders are going to deposit with them.

--------------

As to backtesting so I used backtesting on MT4 just to understand the following: EA works or not (by openning and closing the trades). If I wanted to have some idea about profit from the EA so I am trading it on demo with my broker for at least 3 months. But it is my choice: I believe in trading, not backtesting (in case of metatrader 4 for example).

If someone told me: "my EA is profitable because I backtested it with good results" so my question will be the following: "which broker?". Because everry broker is having different quotes/data (sslightly different). Example: well-known iTrend indicator is having differemnt value for different brokers, next example: famous Brainwashing EA - different results for different brokers.

That is why I like MT5: I do not need to trade by EA on demo for 3 months just to understand: EA may be profitable or not for that past period of time. I just need to backtest EA on MT5 with 'every tick based on real tick' mode with selected broker.

Well with my ea I overall in growth from last two years it was not days process it took long  before my ea I used my strategy as manual for 3 years and  making ea took a long time as it's difficult what we thinks ea will do same or how we did manual trade ea will do same but it took 7 months forward test to adjust ea completely like I had manual trade. So after 2 years with ea now I planned to open a MT5 base account but it taking long time to make sure all going well. My ea already checked after a two years trade via MT5 backtest I'm checking broker so confused when results changed.

On demo account no doubt my strategy can do same as backtest I checked it 3 months but  Demo is nothing broker don't need to send orders to LP so no slippage so demo can not tell the real story. I trust only on real account results  but backtest can help a lot for preparing. I always assumed backtest and forward test have 30% different in results so I tried to win 90% in backtest meaning winning 60% in forward test with 60% winning overall we can see growth.

You experienced in your own way  I understand it as trading is totally individual war every mind must need to learn by own as its real time decision system. Hope I can set next 3 years plan on MT5 as its fifth generation platform.