FPC
When the Search function comes back online here, you can look through many threads on backtesting variations etc, etc
But you will still be left with the differences between even single-broker history data and live trading on that broker
Performance on any two brokers can vary enormously - even if they are the same type, i.e. both Dealing-desk, both STP, or both ECN
You will even get differences between live accounts of equal type/amount, with the same EA on the same broker..
So just get used to the idea that back-testing & even live trading is not an exact science
Maybe back-testing will will be less inexact in MT5 but there will always be a difference to live
My 2c worth
-BB-
BB:
I must ask then, how much weight does back testing and optimization carry in your analysis of a trading strategy? Live trading results no doubt carry the most, but getting these results is such a timely process....
Thank you,
FPC
Get used to it!
Its a tough business, merciless even, so lets not rush in!
I would normally only code systems that have been manually tested as profitable by myself or my associates
With backtesting you are looking for "whats the worst that might happen"
With optimization you are looking for changes in system behaviour in high/low volatility, trends compared to ranges, are Fridays significantly different to other days, etc, etc
Things that would be tricky to spot when watching just one live trade at a time
You dont want to optimize to the last pip of StopLoss, but you might find that Buys should be weighted more in uptrends, TakeProfit might be better linked to Daily ATR than a fixed amount, etc, etc
This is all why I have quite a few PCs..!
Good Luck
-BB-
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Hello all:
I have a wonderful EA I wrote but I'm having some problems:
I’m doing some thorough testing and I’ve found that running my same EA on the same type of platform (PFG MT4) demo account, I get conflicting trades. One platform will execute an order (when it should) and with all the same settings, the other platform will not trade. I have triple checked the journal and there is nothing, absolutely nothing.
Also, from platform to platform, (PFG to FXDD) after downloading the data (which I thought it all comes from metaquotes) I get varying back tests. Can anyone explain either of these phenomenons to me?