| Backtests
All the following backtests were performed using tick-by-tick data with real spreads instead of the commonly used MT4's 90% modelling quality and/or fixed spreads, which is insufficient and shows completely irrelevant results (especially for scalpers). Because of the MT4 2GB data file restrictions and lack of multi-currency backtest options, we had to backtest the software part by part and then compound the statements to PDF (the original MT4 statements are available in the backtest section for reference).
Please note that the software is backtested since April 2007. It is a common mistake to backtest scalping EA's on only 90% modelling quality, as such quality is insufficient and gives completely irrelevant results, which do not match forward tests and are therefore worthless. All our backtests are based on tick-by-tick data, which are only available since April, 2007.
Although it is usually true that backtesting is worthless if used alone as the only "proof" of results, it is extremely important if used in comparison with forward test results. If forward test results match the backtest applied on the corresponding period of time, it means that the particular backtest is a completely valid indicator of the software's performance (provided the time period was long enough - say at least 20 trades were taken). This can be only achieved by testing on tick-by-tick data and on spreads reached in reality, including commissions.
Unfortunately, as far as we are aware, we are the first and so far the only ones to provide backtests on tick-by-tick modelling quality with real spreads - all other backtests published in other presentations nowadays use 90% modelling quality and/or unrealistically low spreads and most of them turn into losers when backtested under real conditions and/or forward tested, which damages the reputation of backtesting, although it is a priceless tool if performed correctly.
|
