[Back-test][MT4] XMT Scalper

Share your Tickstory back-tests for Expert Advisers and Strategies
Post Reply
tickstory
Posts: 5075
Joined: Sun Jan 06, 2013 12:27 am

[Back-test][MT4] XMT Scalper

Post by tickstory »

Description:
XMT Scalper was a classic EA that was being developed in a community effort on a (now-defunct) user forum. It was based on a popular scalper called "Million Dollar Pips" which attempted to profit from short-term movements as price breached a trading channel. It was designed to work on any instrument that exhibited some volatility.

Before we begin - a standard Disclaimer:
This case study is for demonstrating how to use Tickstory to test your own EAs. Past performance is no guarantee of future results. We do not make any claim or guarantee about this EA. Use it live at your own risk! The EA has no technical support.

Test Requirements:
- XMT Scalper EA (see attached)
- Metatrader 4
- Tickstory & Advanced Tester

Test setup:
- Symbols: EURUSD
- Timeframe: 15-minute
- Test data between 2023-06-19 - 2024-06-19 with "embed spread" turned on
- EA Settings: The EA has a range of settings to tweak but for the purposes of our test, we won't deviate too far from the standard settings aside from changing profit targets and stop loss.

Basic steps to conduct the test:
- Export EURUSD data from Tickstory with 'embed spread' turned on. To minimize disk-space usage, only M1, M5 and M15 FXT time-frames are exported (we only use M15 in this test though).
image.png
- Launch Advanced Tester and enable 'variable spread' as shown in the screenshot.
image.png
- Select the XMT Scalper EA in the MT4 Strategy tester, EURUSD, 15-minute time-frame and use entire date range
- Begin the test!

Test Results:
Equity Graph
Equity Graph
Back-test Report
Back-test Report
Conclusion:
Like with many scalping EAs, the equity curve shows that it essentially treads water regardless of how you adjust the profit target and stop-loss. Sometimes it has a bout of success but then goes through a period of draw-drawn where it loses any gains. Inevitably it whittles down your balance to zero since spreads and commission are always working against you.

Before the advent of Tickstory, many users were encouraged by the results of these types of EAs since they were testing with 1-minute input data and fixed spreads. This gave the impression that profit targets were being hit sooner than they would in real trading.

When performing a more accurate back-test using Tickstory's quality tick data with variable spreads, it is more apparent that this type of EA is difficult to get right and requires a more compelling strategy to beat the market. We didn't even need to use the Advanced Tester's slippage simulation because it was clear that this strategy would fair even worse with it activated.

Feel free to try out the EA, contribute to some improvements, continue the discussion and/or share settings files. What will it take to make this strategy profitable?

Download link:

[You must be logged in to view this attachment.]

Post Reply