Metatrader expert advisors dissected.
PhiBase Synergy live forward test
Trading style: attempts to capture large market moves based on several chart patterns; uses a trailing stop.
Currency pairs: EURUSD, GBPUSD, USDCHF
Timeframe: H1
EA price: $350
License: 2 live accounts, 2 demo accounts
NFA compliance: yes
Refund policy: none
Read more at the PhiBase Synergy website
Buy PhiBase Synergy
Note: the EA changed its name from PhiBase Pro to PhiBase Synergy when v3.2 was released during the 2nd half of July 2012.
Birt’s forward test
Settings: default (Max_Allocation_Per_Trade 10.0, Geometrical_MM true)
Started: 15.02.2012
Broker: PrivateFx
Account type: live, micro
Starting balance: $300
Currency pairs: EURUSD
Current version: 4.03
Official accounts
Settings: default
Started: 13.06.2012
Broker: FxPro
Account type: live
Starting balance: $3000
Currency pairs: EURUSD, USDCHF
Backtests
All backtests were performed with v1.23. Newer versions might display better performance.
EURUSD
History center data, spread 1.5
PhiBase Pro v1.23 1999-2012 EURUSD history center data backtest, spread 1.5, default settings
| Symbol | EURUSD (Euro vs US Dollar) | ||||
| Period | 1 Hour (H1) 1999.01.04 10:00 - 2012.02.29 23:00 | ||||
| Model | Every tick (the most precise method based on all available least timeframes) | ||||
| Parameters | Activation_Code=; Variable_LotSize=5; Fixed_LotSize=0; MagicNumber=99118260; Targetlevel_1=50; TrailSL_1=30; Targetlevel_2=100; TrailSL_2=50; Higher_Band=150; Lower_Band=100; | ||||
| Bars in test | 82757 | Ticks modelled | 66761817 | Modelling quality | 90.00% |
| Mismatched charts errors | 0 | ||||
| Initial deposit | 10000.00 | ||||
| Total net profit | 41772.69 | Gross profit | 163126.35 | Gross loss | -121353.67 |
| Profit factor | 1.34 | Expected payoff | 29.21 | ||
| Absolute drawdown | 930.25 | Maximal drawdown | 5502.40 (10.31%) | Relative drawdown | 25.18% (3432.72) |
| Total trades | 1430 | Short positions (won %) | 597 (65.49%) | Long positions (won %) | 833 (68.43%) |
| Profit trades (% of total) | 961 (67.20%) | Loss trades (% of total) | 469 (32.80%) | ||
| Largest | profit trade | 2816.82 | loss trade | -1928.32 | |
| Average | profit trade | 169.75 | loss trade | -258.75 | |
| Maximum | consecutive wins (profit in money) | 21 (2867.16) | consecutive losses (loss in money) | 7 (-1900.21) | |
| Maximal | consecutive profit (count of wins) | 7223.44 (16) | consecutive loss (count of losses) | -2958.33 (3) | |
| Average | consecutive wins | 3 | consecutive losses | 2 | |
Tick data, real spread
PhiBase Pro v1.23 2007-2012 EURUSD tick data backtest, real spread, default settings
| Symbol | EURUSD (Euro vs US Dollar) | ||||
| Period | 1 Hour (H1) 2007.03.30 16:01 - 2012.01.31 12:59 (2007.07.01 - 1970.01.01) | ||||
| Model | Every tick (the most precise method based on all available least timeframes) | ||||
| Parameters | Activation_Code=; Variable_LotSize=5; Fixed_LotSize=0; MagicNumber=99118260; Targetlevel_1=50; TrailSL_1=30; Targetlevel_2=100; TrailSL_2=50; Higher_Band=150; Lower_Band=100; | ||||
| Bars in test | 30196 | Ticks modelled | 67882776 | Modelling quality | 99.00% |
| Mismatched charts errors | 0 | ||||
| Initial deposit | 10000.00 | ||||
| Total net profit | 33904.90 | Gross profit | 101475.75 | Gross loss | -67570.84 |
| Profit factor | 1.50 | Expected payoff | 54.69 | ||
| Absolute drawdown | 629.72 | Maximal drawdown | 4943.36 (10.18%) | Relative drawdown | 16.41% (2477.73) |
| Total trades | 620 | Short positions (won %) | 289 (71.63%) | Long positions (won %) | 331 (76.13%) |
| Profit trades (% of total) | 459 (74.03%) | Loss trades (% of total) | 161 (25.97%) | ||
| Largest | profit trade | 2517.22 | loss trade | -1447.21 | |
| Average | profit trade | 221.08 | loss trade | -419.69 | |
| Maximum | consecutive wins (profit in money) | 20 (2200.44) | consecutive losses (loss in money) | 7 (-614.64) | |
| Maximal | consecutive profit (count of wins) | 6021.01 (16) | consecutive loss (count of losses) | -2452.11 (3) | |
| Average | consecutive wins | 4 | consecutive losses | 1 | |
The EURUSD backtest is very good as you can see in the charts above, but the other two pairs aren’t keeping up the same pace. The vendor mentioned that the standalone strategies for GBPUSD and USDCHF will not be profitable when the parameters aren’t updated based on the price action, which I understand will happen 3-4 times per year. Quote from the vendor regarding the two additional pairs (GBPUSD & USDCHF):
It is optional for the users to choose running these additional pairs. We do not recommend running these pairs for stand-alone trading using the recommended parameters.
GBPUSD
History center data, spread 2.0
PhiBase Pro v1.23 1999-2012 GBPUSD history center data backtest, spread 2.0, GBPUSD settings
| Symbol | GBPUSD (Great Britain Pound vs US Dollar) | ||||
| Period | 1 Hour (H1) 1999.01.04 07:00 - 2012.02.29 23:00 | ||||
| Model | Every tick (the most precise method based on all available least timeframes) | ||||
| Parameters | Activation_Code=; Variable_LotSize=5; Fixed_LotSize=0; MagicNumber=99118260; Targetlevel_1=0; TrailSL_1=0; Targetlevel_2=0; TrailSL_2=0; Higher_Band=500; Lower_Band=150; | ||||
| Bars in test | 82392 | Ticks modelled | 65477575 | Modelling quality | 90.00% |
| Mismatched charts errors | 0 | ||||
| Initial deposit | 10000.00 | ||||
| Total net profit | -7460.37 | Gross profit | 78805.38 | Gross loss | -86265.75 |
| Profit factor | 0.91 | Expected payoff | -3.72 | ||
| Absolute drawdown | 7875.35 | Maximal drawdown | 8244.26 (79.51%) | Relative drawdown | 79.51% (8244.26) |
| Total trades | 2007 | Short positions (won %) | 685 (59.71%) | Long positions (won %) | 1322 (59.68%) |
| Profit trades (% of total) | 1198 (59.69%) | Loss trades (% of total) | 809 (40.31%) | ||
| Largest | profit trade | 679.01 | loss trade | -384.95 | |
| Average | profit trade | 65.78 | loss trade | -106.63 | |
| Maximum | consecutive wins (profit in money) | 15 (817.36) | consecutive losses (loss in money) | 10 (-402.19) | |
| Maximal | consecutive profit (count of wins) | 1299.47 (10) | consecutive loss (count of losses) | -971.74 (6) | |
| Average | consecutive wins | 2 | consecutive losses | 2 | |
Tick data, real spread
PhiBase Pro v1.23 2007-2012 GBPUSD tick data backtest, real spread, GBPUSD settings
| Symbol | GBPUSD (Great Britain Pound vs US Dollar) | ||||
| Period | 1 Hour (H1) 2007.03.30 16:01 - 2012.01.31 08:59 (2007.07.01 - 1970.01.01) | ||||
| Model | Every tick (the most precise method based on all available least timeframes) | ||||
| Parameters | Activation_Code=; Variable_LotSize=5; Fixed_LotSize=0; MagicNumber=99118260; Targetlevel_1=0; TrailSL_1=0; Targetlevel_2=0; TrailSL_2=0; Higher_Band=500; Lower_Band=150; | ||||
| Bars in test | 30191 | Ticks modelled | 62667236 | Modelling quality | 99.00% |
| Mismatched charts errors | 0 | ||||
| Initial deposit | 10000.00 | ||||
| Total net profit | -2393.17 | Gross profit | 53321.40 | Gross loss | -55714.57 |
| Profit factor | 0.96 | Expected payoff | -3.02 | ||
| Absolute drawdown | 3444.91 | Maximal drawdown | 5449.35 (45.39%) | Relative drawdown | 45.39% (5449.35) |
| Total trades | 792 | Short positions (won %) | 258 (63.57%) | Long positions (won %) | 534 (59.74%) |
| Profit trades (% of total) | 483 (60.98%) | Loss trades (% of total) | 309 (39.02%) | ||
| Largest | profit trade | 854.12 | loss trade | -935.48 | |
| Average | profit trade | 110.40 | loss trade | -180.31 | |
| Maximum | consecutive wins (profit in money) | 12 (1361.77) | consecutive losses (loss in money) | 8 (-1288.43) | |
| Maximal | consecutive profit (count of wins) | 1753.19 (5) | consecutive loss (count of losses) | -2055.80 (4) | |
| Average | consecutive wins | 2 | consecutive losses | 2 | |
USDCHF
History center data, spread 2.0
PhiBase Pro v1.23 1999-2012 USDCHF history center data backtest, spread 2.0, USDCHF settings
| Symbol | USDCHF (US Dollar vs Swiss Franc) | ||||
| Period | 1 Hour (H1) 1999.01.04 08:00 - 2012.02.29 23:00 | ||||
| Model | Every tick (the most precise method based on all available least timeframes) | ||||
| Parameters | Activation_Code=; Variable_LotSize=5; Fixed_LotSize=0; MagicNumber=99118260; Targetlevel_1=0; TrailSL_1=0; Targetlevel_2=0; TrailSL_2=0; Higher_Band=0; Lower_Band=0; | ||||
| Bars in test | 82631 | Ticks modelled | 61411593 | Modelling quality | 90.00% |
| Mismatched charts errors | 0 | ||||
| Initial deposit | 10000.00 | ||||
| Total net profit | 15881.67 | Gross profit | 153720.01 | Gross loss | -137838.35 |
| Profit factor | 1.12 | Expected payoff | 7.87 | ||
| Absolute drawdown | 1111.81 | Maximal drawdown | 4989.35 (35.30%) | Relative drawdown | 35.30% (4989.35) |
| Total trades | 2019 | Short positions (won %) | 1012 (57.61%) | Long positions (won %) | 1007 (57.99%) |
| Profit trades (% of total) | 1167 (57.80%) | Loss trades (% of total) | 852 (42.20%) | ||
| Largest | profit trade | 2986.28 | loss trade | -854.80 | |
| Average | profit trade | 131.72 | loss trade | -161.78 | |
| Maximum | consecutive wins (profit in money) | 11 (1220.06) | consecutive losses (loss in money) | 6 (-691.47) | |
| Maximal | consecutive profit (count of wins) | 5105.76 (6) | consecutive loss (count of losses) | -1543.77 (5) | |
| Average | consecutive wins | 2 | consecutive losses | 2 | |
Tick data, real spread
PhiBase Pro v1.23 2007-2012 USDCHF tick data backtest, real spread, USDCHF settings
| Symbol | USDCHF (US Dollar vs Swiss Franc) | ||||
| Period | 1 Hour (H1) 2007.03.30 16:01 - 2012.01.31 12:59 (2007.07.01 - 1970.01.01) | ||||
| Model | Every tick (the most precise method based on all available least timeframes) | ||||
| Parameters | Activation_Code=; Variable_LotSize=5; Fixed_LotSize=0; MagicNumber=99118260; Targetlevel_1=0; TrailSL_1=0; Targetlevel_2=0; TrailSL_2=0; Higher_Band=0; Lower_Band=0; | ||||
| Bars in test | 30182 | Ticks modelled | 53138873 | Modelling quality | 99.00% |
| Mismatched charts errors | 0 | ||||
| Initial deposit | 10000.00 | ||||
| Total net profit | 10000.90 | Gross profit | 54067.42 | Gross loss | -44066.52 |
| Profit factor | 1.23 | Expected payoff | 12.84 | ||
| Absolute drawdown | 3458.06 | Maximal drawdown | 3462.97 (34.61%) | Relative drawdown | 34.61% (3462.97) |
| Total trades | 779 | Short positions (won %) | 403 (62.53%) | Long positions (won %) | 376 (59.04%) |
| Profit trades (% of total) | 474 (60.85%) | Loss trades (% of total) | 305 (39.15%) | ||
| Largest | profit trade | 2513.77 | loss trade | -699.90 | |
| Average | profit trade | 114.07 | loss trade | -144.48 | |
| Maximum | consecutive wins (profit in money) | 11 (1029.03) | consecutive losses (loss in money) | 5 (-1343.81) | |
| Maximal | consecutive profit (count of wins) | 3948.47 (5) | consecutive loss (count of losses) | -1343.81 (5) | |
| Average | consecutive wins | 2 | consecutive losses | 2 | |
-
-
#3 written by Murray February 15, 2012 (1 year ago)
-
#4 written by Julien February 23, 2012 (1 year ago)
Murray,
I did that test for EURUSD 2011, shifted 28 years back… The result was the same as with non shifted data… So no hard coded dates probably, at least for 2011…
Another evidence: the EA was a looser from january to june 2011 (which you can also see from official backtests)…
It is worth saying that I was very surprised to see a bunch of indicators appear after the backtest… Although the EA is supposed to trade chart patterns, it actually uses more indies than most of my indicator-based EAs !!-
#5 written by birt February 24, 2012 (1 year ago)
-
-
-
#7 written by Alex February 24, 2012 (1 year ago)
-
#9 written by K.L.A. February 29, 2012 (1 year ago)
-
#11 written by eaviews February 29, 2012 (1 year ago)
Extracted from Vendor’s newsletter on FAQ :
——————————————————————-
PhiBase PRO is based on Chart Patterns – Then what is the role of
indicators?The chart patterns are normally identified using candles – These work best for large time frames like H4, D1, etc. In H1 charts the noise is very
high and it is possible to see or not see the same patterns which seem to
be clear on the higher time frames. PhiBase was developed for H1 time frame to minimize GMT related differences. A range of very slow indicators are used to provide inputs to know what is happening on the larger timeframe (H4). These indicator along with the H1 bars enable the EA to identify a set of points which we use as a pattern bounding box. Instead of relying on absolute values of bars and indicator values, the normalization is used to eliminate time-price range related info to produce simplified set of 8 point pattern within the bounding box. So the indicators being used will help in minimizing broker related differences.Difference between brokers cannot be totally eliminated. Most trades will
be similar when considered over longer periods of time.
——————————————————————- -
#12 written by Karan March 4, 2012 (1 year ago)
-
#13 written by birt March 5, 2012 (1 year ago)
-
-
#14 written by dash March 5, 2012 (1 year ago)
where is the backtest brit? can this one be trusted? http://www.myfxbook.com/strategies/phibase-v123-5-year-tick/13229
-
#16 written by John March 6, 2012 (1 year ago)
-
#18 written by Mike March 7, 2012 (1 year ago)
Hi – been running phibase for two weeks now. I am running all 3 pairs – same like the vendor reference account. they say eurusd is recommended pair. GU and UC are for risk diversification only and are to be used with specific settings they give. No complaints, working great and good support. From the trades so far it looks like it is indeed able to identify some patterns – i could make out the double top and a few more – but would have been impossible to identify them if the trade was not taken – all were pretty tough calls. must learn to sit back and leave ea 2 trade – trying to run tickdata backtest.
-
#20 written by Mike March 30, 2012 (1 year ago)
I started 3rd week of feb – equity up 18% with gain of 507 pips. 220 pips booked this week. Except for two news related losses, the Ea has been exceptionally great. Birts index is 1.39 as of today and its ranked 5th in birts top 30 list – fully agree with your ranking system, Birt…. its one of the best and most logical ways to rate EAs.
I like the strategy and support is fanastic.
-
#22 written by Mark April 27, 2012 (1 year ago)
-
#23 written by birt April 30, 2012 (1 year ago)
As specified above, I use v2.2. I’m not sure if the vendor is using the same version but it would appear so given the account names. Anyway, it trades slightly different from a broker to another as you can see in the forward tests posted above and that’s probably the reason why that particular trade doesn’t show up in my account. Most likely, there are also other trades that are missing or extra if you look closer.
-
-
#24 written by Mike April 30, 2012 (1 year ago)
Hi Birt, We were provided with the latest build – 20120422 (same v2.2 but the problem with instant trade open-close corrected). It looks like you have the v2.2 installed but not the latest one. The latest version will display a EA message that reads like “Version 2.2 build date 20120422″ in green color!
You have 3 trades that closed within the same minute – I am sure you need to get the update – I always find the vendors email in my spam for some reason! May be you may find their update email there
-
#29 written by Mike June 19, 2012 (11 months ago)
Hi Birt, Your test are great. It shows how running the EA consistently and without interference can produce good results and also allow the EA to recover from DD’s/
I need a little bit of clarification on the type of broker account you are using – your tests shows that the phibase trade sizes are around 0.05 micro lots on a capital of just $ 300 and I suppose you are still using v2.2
I am trying to run the EA on a second account with a capital of $1000, but even with a max allocation of 5, the EA calculates the trade size as 0.01 microlot. I checked up with the vendor and hey confirmed that it is correct.
I am not able to understand the trade sizes being taken on your account. Also please let use know the broker details like leverage, minimum lot size allowed and what the lotsize means in terms of pips-$ movement.
Do let us know when you switch over to v3.0 – Would you consider running USDCHF pair along side (or on a separate account?)
Keep up the great work!
-
#30 written by birt June 19, 2012 (11 months ago)
I updated to v3.0 today.
The major difference for PrivateFX is that the contract size (lot size) is 10000 instead of 100000 as it is on most brokers. This results in a pip value that is exactly 10 times smaller than on most brokers.
Finally, regarding USDCHF, I’m not in a rush to add it. It has performed rather poorly on the vendor’s accounts. Sure, I might change my mind if that changes in the future.
-
-
#31 written by Mike July 23, 2012 (10 months ago)
synergy v3.2 was delivered today.
Birt, please try to update with this version. and any plans to reveiw this? Its done very well on my account – up by over 70% in 5 months using risk 10 – had a max dd of close to 25% in May, but it recovered quite well and blasted in to green by end of June. Has been the best one EA I have used so far and considering to add to additional accounts now – your opinions will ofcourse help in going about this more confidently.
-
#32 written by birt September 12, 2012 (8 months ago)
I updated to v3.2 some 3 days after your comment but forgot to reply.
While it might be a good candidate for a review, I have a whole host of issues on my plate, issues that I must take care of before I even think of taking the time to write a full review so I don’t expect it to happen anytime soon. Really wish I could purchase some time and energy to do all the things I want to do…
-
-
#35 written by phibase September 23, 2012 (7 months ago)
The synergy results since 1st August are summarized below:
Official reference on our website
Birt’s : EURUSD : + 105 pips (real. runs only EURUSD)
FxOpen : EURUSD : + 806 pips and USDCHF : +396 pips > Net +1202 pips (Real)
FxOpen : EURUSD : + 581 pips and USDCHF : +415 pips > Net + 996 pips (Demo)
FXPRO : EURUSD : + 271 pips and USDCHF : +894 pips > Net +1165 pips (Demo)User1 : EURUSD : + 855 pips and USDCHF : +469 pips > Net +1324 pips (Demo)
User2 : EURUSD : + 937 pips and USDCHF : +496 pips > Net +1433 pips (Demo)
User3 : EURUSD : + 1266 pips (real. runs only EURUSD)In general synergy has done well on most broker accounts. Most user accounts have better results that our own reference accounts. Birt real reference account missed two EURUSD trade entries due to which it has much lower gains. One of the missing trade was also common between other accounts and can be attributed to broker feeds. The other missing trade was most probably due to connection failure or some issue which will need to be identified.
-
#38 written by phibase September 26, 2012 (7 months ago)
One of our members has reported total gains of +448 on EURUSD and CHF combined. This is the feedback from live forward test on insta forex real account with spread of 3 pips.
We find that this was much below the average gains seen on most of the other accounts we are tracking – the average total gains on both pairs combined was above 1000 pips from August till Mid September
PhiBase Synergy is designed to be broker independent and trades are expected to be similar across all brokers. We are not sure if this is a single case and are inviting feedback from all members so that we can correlate results better.
-
#39 written by Ronald October 1, 2012 (7 months ago)
Hi there,
after doing some more reading, I decided to let go of Fap Turbo and buy Phibase v3.2, which I did last week. So, it’s up and running as from this moment… While reading over your Phibase review again, I was wondering why the online manual only mentions two pairs to trade: eurusd and usdchf(strangely enough, the downloaded zip-file doesn’t inlude a user guide), while no word about gbpusd – enless I missed it. It does explain that the ea property ‘pair’ has to be set to ’1′ for eurusd and ’2′ for ‘usdchf’, so if there was mentioning of a thired pair, it would have said to set the ‘pair’ to ’3′ for that one…
Anyway, since the market has only opened up a few hours ago, the ea hasn’t taken any trades yet – also guess that it has perform some calculations before entering its first trade(s). Hope I did the write thing by returning FT56 (and the included Ichimoku ea) to its makers – according to them, I’m making a big mistake, of course.
Thanks for the welt of information you present on your site, Birt!
Regards,
Ronald
-
#40 written by phibase October 1, 2012 (7 months ago)
Some part of the information on Birt’s site will need to be updated.
PhiBase Synergy was originally released in February as PhiBase Pro (v1.23). The back tests shown on Birt’s website were also run using the first version.
After this we updated the EA to v2.2 and then to v3.2 – With in the release of v3.2, the EA was renamed to PhiBase Synergy.
V1.23 was recommended for EURUSD – Additional pairs USDCHF and GBPUSD could be run optionally to provide portfolio diversification. These two pairs were not meant for stand alone trading. V1.23 used band parameters which was used to correlate the price action between EURUSD to these pairs. These band parameters was updated as per the EURUSD price action and our research team initially was to provide the technical analysis for this. However, trading the 3 pairs using this correlation/band parameters was increasing the risk when the EA was undergoing normal DD on the EURUSD. Our development team dropped the strategy of using band parameters and also decided to drop GBPUSD.
Since the launch of v3.2, EURUSD is the recommended pair. USDCHF can also be traded additionally (optional) since it is highly correlated to EURUSD and price action is very similar. We find that the EA is suitable to trade USDCHF to be suitable for stand alone trading.
-
#44 written by phibase January 3, 2013 (4 months ago)
Hi Birt,
The PhiBase Reference test account seems to have some issues. Can you please have a look at it at your end for any EA messages/expert or journal log errors. The EA has been continuously missing all the new trades. The current basket (entries between 10th to 18th Dec) which is open in this account were closed out on all other accounts on 20th December. After that there were 10 new trades. Our earlier email to you in this regard does not seem to have reached you (ends up in spam some times).
Hope you will be able to look into this.
Thanks and best regards,
Megan – PhiBase Tech_Support
-
#45 written by birt January 3, 2013 (4 months ago)
The EA appeared to be in control of the trades – it says on the chart: 1.30182 will trigger hidden stop loss and 1.32182 will trigger trailing SL. The issue was brought to my attention on myfxbook a few days ago and I figured I would let the EA handle it since it appeared to have things under control. There were some 2h of server downtime on 19.12 which might be what lead to the whole issue. In any case, I have now closed the trades manually and restarted the terminal. Unfortunately, the manual close resulted in a rather large loss but there’s nothing I can do about that.
As for the email you sent me, I don’t seem to have received it. I even searched the spam folder now that you mentioned the email but it wasn’t there either.
-
-
#46 written by phibase January 3, 2013 (4 months ago)
Thanks, Birt – for the swift response and closing out the trades. Atleast we can have the account back in line with other accounts to help provide reference.
A few hours of downtime will not cause this problem. Just to be sure, please do have a look at the expert and journal tabs for any error messages. We can rule out any other issues that could have affected trading on this account.
We will check up with our hosting service provider if there is any reason why our emails are not reaching you. Some servers block out our emails originating from the whole network even if we are not the source of spam. We will email you as soon as this is rectified.
-
#47 written by REH February 28, 2013 (2 months ago)
-
#48 written by birt February 28, 2013 (2 months ago)
-
-
- Comment Feed for this Post
You can run on 2 account, with 1 license.