What everyone wants to know! Q&A for "Pips_Miner_EA"
Dr. Neko's latest creation is GBPUSD high-frequency trading scalping!
How does the performance compare to the previous EAs?!
【Pips_Miner_EA Overview】
Currency pair: [GBP/USD]
Trading style: [Scalping]
Maximum number of positions: 3 Others: configurable below 10
Timeframe used: M15
Maximum stop loss: 80
Take profit: 8
Despite the short forward period, more than 200 people have already purchased within one week of release!
【Backtest Analysis】
2008.01.01-2017.12.31
Spread 1.5
0.1 lot fixed (×3 positions)
Net profit +1,180,000
Maximum drawdown 150,000
Total trades 6,820 (annual average 682)
Win rate 91%
This is the result.
The maximum drawdown per 0.1 lot is 150,000 yen, and per 0.1 lot for GBPUSD the
margin is 75,000 yen, so
225,000 yen + (150,000 yen × 2) = 525,000 yen
Thus, you can operate from around 500,000 yen.
With 1,000,000 yen, perhaps 0.2 lots would be suitable?
The expected average annual return in that case is＋23.6％.
【Annual Profit/Loss】
【Monthly Profit/Loss】
Typically, April and June tend to be less favorable (loss-prone).
【Parameter Highlights】
■Boost (Trading frequency)
Boost (Trading frequency setting) = 0; in the range 0–10, closer to 0 means higher trading frequency, closer to 10 means fewer trades with more selective entries.
■MM Setting
MM percent is the risk percentage that automatically adjusts the lot size so that the loss per position equals 1% of the account balance when compounding is used.
Even with default compounding, the risk is low but profits are also reduced, so
you should definitely set this.
■ Max_lot_size
Maximum trade lot size
Defaults to 1 (100,000 units).
If compounding for several years, the entry lot may quickly reach 100,000 units, so adjust the max lot size to fit your account funds.
【Compare Performance with Historical EAs】
Pips_Miner_EA is characterized by a high trading frequency, but how much does it outperform other EAs?
We compared it using official backtest data.
All use 0.1 lots (Fubuki defaults to 0.05 lot).
When comparing annual average trades, it certainly has by far the highest trading frequency!
If you only look at maximum drawdown, Scal_USDJPY with 2 positions has the lowest drawdown; an EA you might want to use with larger lots.
In terms of expected profit (net profit ÷ max drawdown),Pips_Miner_EA has the best performance, so for those who want high-frequency trading with high profits,
this is a very welcome new release.
The most popular "Scal_USDJPY" has different currency pair and profit curve, so it is also recommended for portfolio management.
（Pips_miner_EA and Scal_USDJPY profit curves comparison）
Pips_miner_EA can have drawdown concerns, but Scal_USDJPY shows a slow but steady upward trend.
【Recommended Boost Values & Optimizing Compounding Percentage】
Boost value: 0–10
We performed a simple optimization at control points.
The period is 2013.01.01–2017.12.31 (5 years)
Maximum lot size changed from 1.0 to 10 (1,000,000 units)
（※Profits are in dollars）
As a projection, we thought Boost high × MM high would yield higher profits with high win rate × high leverage, but
the result was Boost1 × MM10 giving the highest profit amount. The maximum drawdown was in the low 30% range, so
it's clear that raising the MM percentage is okay.
（Note: backtests cover 5 years; results may differ for longer periods）
Surprisingly, Boost=0 and MM=10 did not yield very high profits
(though +500% over 5 years is still plenty)
The results including the number of trades are shown below:
The above results are from control-point optimization, so before live operation please run a long-term backtest on all ticks.
【Operational Cautions】
Given that the take-profit is 8 pips,narrower spreads are more favorable.
Regardless of domestic or overseas FX brokers, before trading, be sure to check how wide the GBPUSD spread is.
If the spread is under 1.5 pips, you can expect backtest-like performance!
Realize high-frequency trading × high profits!
GBPUSD Scalping
