Operating since May 2015, with an annual return over 23%: CounterQ Span Model and RSI-based, easy-to-control EA

Recently, the EA (FX automated trading software) whose purchases by customers have been notably increasingCounterQ has drawn a lot of attention.

We also conducted a phone interview with the developer, so I would like to introduce the situation as well.

　CounterQ’s logic is, as described on the EA product page (CounterQ) and in the included manual, that it trades USD/JPY on a 5-minute chart with a single position (only one position), with a take-profit value (win) of 25 pips and a stop-loss setting value (loss) of 120 pips, uses a Span Model and RSI, and focuses on a certain period of range-bound price action to avoid trend reversals and the trend market itself as much as possible.

The Span Model excels at distinguishing between range-bound and trend markets, so it is designed not to enter during trend reversals or while a trend is developing.

Also, RSI is an oscillator-type technical analysis used to determine overbought and oversold conditions, so it is thought that trading is halted by aligning the timing when the Span Model detects a trend with the RSI's overbought/oversold timing.

Stop-loss setting value(loss) is 120 pips, which assumes entries are made in a ranging market and that even when moving against the entry, it will return. And the take-profit value(win) of 25 pips is designed to secure a win.

This design of trading only in a range-bound market is reflected in the actual forward test by a high win rate of 70.16%, as shown in the figure below.

Forward Test Results from May 11, 2015

The developer Okon says that this is an “easy-to-control EA for beginners.” The reason is that, due to the high win rate, there are few consecutive losses that reach the large stop-loss setting value (loss), so people who started using the EA or have relatively little trading experience are less likely to stop operating due to fear of losses, which is probably the basis for this claim.

However, when you actually look at the profit-and-loss curve below, from July to November 2017 there was a drawdown of 17.45% (38,050 yen), and surely at that timing many would stop operating.

Forward test P/L curve from May 11, 2015

　During 2017, there were sudden trends that presumably caused CounterQ's ability to distinguish between trend markets and range markets to fail.

In such timing, stopping is prudent, but according to recent conversations with EA Azarashi, White Bear's Rakko-san, and Flashes' Trader Kaibe-san, they run the EA with large lots that do not match the margin, which makes them feel they must stop; otherwise, continuing to operate without stopping, an EA with excellent logic would ultimately be profitable due to the ongoing sense of yearly market risk, I recall.

I thought it would be good if more investors used such operating practices.

That said, the developer Okon emphasizes on the EA's product page and manual that “by avoiding known event risks in advance, it shows more stability than a demo forward,” and the parameters include

Weekend Stop※ false: Weekend trading (true/false)

Stop̲Day̲Of̲Week 5     :Specified day of the week (0: Sunday～6: Saturday)

Stop̲Hours 0(時) :Trade stop time (server time)

DayTrade̲Stop  false :Intraday entry restriction (true/false)

startTime 0:00  :Start time (server time)

EndTime 23:50 :End time (server time)

They have implemented these finely adjustable stop features.

Setting parameters every weekend while considering next week's events and then starting operation at the beginning of the week could yield even better results, I think.

As above, Trader Kaibe-san also claims this and posts daily EA stop times on Twitter; among EA users, there may be an increasing number of experienced discretionary traders who have found that stopping operation can further improve performance.

Written by Hayakawa