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
Alpari Data, 2004-08-15 thru 2006-09-30.
Alpari Data
Bars in test 3822
Ticks modelled 2783034
Modelling quality 90.00%
Initial deposit 100000.00
Total net profit 54700.42
Gross profit 91539.47
Gross loss -36839.05
Profit factor 2.48
Expected payoff 283.42
Absolute drawdown 0.00
Maximal drawdown 9249.02 (5.99%)
Relative drawdown 5.99% (9249.02)
Total trades 193
Short positions (won %) 81 (76.54%)
Long positions (won %) 112 (72.32%)
Profit trades (% of total) 143 (74.09%)
Loss trades (% of total) 50 (25.91%)
Largest
profit trade 1762.50
loss trade -2667.50
Average
profit trade 640.14
loss trade -736.78
Maximum
consecutive wins (profit in money) 21 (17616.02)
consecutive losses (loss in money) 8 (-2279.07)
Maximal
consecutive profit (count of wins) 17616.02 (21)
consecutive loss (count of losses) -6866.00 (4)
Average
consecutive wins 8
consecutive losses 3
Build 200 Downloaded data. Same dates, same everything. I clicked the Re-calculate button for both tests.
Bars in test 12507
Ticks modelled 4148388
Modelling quality 90.00%
Initial deposit 100000.00
Total net profit 23738.50
Gross profit 86115.50
Gross loss -62377.00
Profit factor 1.38
Expected payoff 119.89
Absolute drawdown 9420.50
Maximal drawdown 20430.00 (18.40%)
Relative drawdown 18.40% (20430.00)
Total trades 198
Short positions (won %) 87 (64.37%)
Long positions (won %) 111 (68.47%)
Profit trades (% of total) 132 (66.67%)
Loss trades (% of total) 66 (33.33%)
Largest
profit trade 1880.00
loss trade -2604.00
Average
profit trade 652.39
loss trade -945.11
Maximum
consecutive wins (profit in money) 17 (11646.50)
consecutive losses (loss in money) 10 (-6990.00)
Maximal
consecutive profit (count of wins) 11656.50 (13)
consecutive loss (count of losses) -12432.00 (7)
Average
consecutive wins 8
consecutive losses 4
drastic differences!
You call that different?!!!
How about those results from build 200, the first using downloaded mt data, and the other uses alpari data. Both terminals were offline so historical data was not tampered with. Both use the same EA with the same parameters and 90% modeling quality data. Also it's same timeframe, same machine and same user (me!!)
Is there something seriously wrong with the tester in this version?
In any case, attached is the EA in case you wanted to check to see if it will have similar results with you!
This is driving me !
Test results 1:
Test results 2:
feedback appreciated!
You call that different?!!!
How about those results from build 200, the first using downloaded mt data, and the other uses alpari data. Both terminals were offline so historical data was not tampered with. Both use the same EA with the same parameters and 90% modeling quality data. Also it's same timeframe, same machine and same user (me!!)
Is there something seriously wrong with the tester in this version?
In any case, attached is the EA in case you wanted to check to see if it will have similar results with you!
This is driving me !
Test results 1:
Test results 2:
feedback appreciated!If you download from the History Center, you get the datas from MetaQuotes : those datas are NOT filtered by any broker, that's why any scalping strategy works very well. But now, no more broker is enough stupid to let you run those strategies. The History Center is simply useless.
Backtest Differences
Hi, i see the same differences with my EAs, but there is an interesting thing:
I forwardtested my EA from 2007-1-12 to 2007-1-13 , having a profit of 5k (usdmore - usdless), then i backtested the EA from 2007-1-12 to 2007-1-13 (same day as forwardtest) first in the build 198 then in the build 201, the results:
Build 198: 14k (three times more then forwardtest!)
Build 201: 600usd (1/10 then forwardtest!)
Forwardtest: 5k
Well, backtests let me a little confused, that one made with the Build 198 isn't correct (!!!) and that other one made with the Build 201 isn't surely correct (!!!), the only way to give a real reliable backtest result is to let the backtest run on real tick movements data (the data can be recorded and downloaded from mt4 like it's done now for m1-data, m5-data, etc..), i know this can take a lot of GigaBytes of historical data, but i hope that metaquotes will think about it soon, if we are trading to seriously make moneys, it isn't some GBytes of data that scary us :P
Be happy!