Erratic results given by the MT4 strategy tester for the same EA
There is quite a lot on this if you look around the posts, but a key reason is that when you start the tester, it takes the latest known spread and applies it equally across all history, so if the live spread has changed from one test to the next, you may get a different result.
I've taken to disconnecting before doing backtests and definately disconnecting before the Friday close if I plan testing over the weekend. I've found with Alpari some absurd speads just at close, effectively preventing realisic testing on a weekend.
I think there may also be issues relating to how backtester allocates tick values given that it doesn't actually hold historical tick data. Others may elaborate.
V
[...] but a key reason is that when you start the tester, it takes the latest known spread and applies it equally across all history [...]
The market properties used for the test can be seen when pressing the 'Symbol properties' button - these are updated from the broker/account the Terminal is currently connected to (updated with each incoming tick).
It's consistent in interpolating ticks (otherwise the tester would have been completely useless...).
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
As a newcomer to automated trading (and this forum) I have worked out a simple 20 SMA / 10 SMA EA for which the MT4 tester has given promising results after several passages through the MT4 tester (on an Alpari MT4 platform): a 9500 euro gain over a period of 6 weeks for an initial deposit of 5000 euros, with 1.00 lots (100k).
I can' t explain why the same EA, in the same tester, on the same plaftorm, gives totally erratic results 24 hours later (-120 euros or less)!!! I can't see what has gone wrong.
Has anybody experienced the same problem and/or has any idea about a possible solution?
Thanks in advance and 'good trading' to you all.