Technical Report

Back Test Vs Live Trade comparison - 3rd, February 2013

Reliability of backtest and its comparison with live forward trading has always been a subject of interest to traders. At Phibase, we conduct all our back tests using high quality Alpari M1 data obtained from a private source. This dataset does not contain holes/gaps seen in metaquotes data. Usually there are variations in OHLC of actual live feeds when compared to back test datasets. But this difference is very high when compared with metaquotes historical data - especially on lower time frames like M1. Good quality Alpari historical data sets (http://asirikuy.com is one of the sources) are more reliable and have minimal differences even when compared at M1 time frame.

Dukas copy tick data is considered to be the most reliable. Metatrader 4 also makes use of the HST files when calculating the indicators. The HST files that match your FXT file MUST be available in the accounts history folder. If not, tick data backtesting will produce unreliable results

Always use a different MT4 platform to backtest using Metaquotes data. Do NOT use the history center to obtain historical data on your live trading platform.

We find that it would be a good practice for traders to conduct backtest on the live feeds obtained on their own broker platforms. Every broker has their own feeds and usually all brokers have M1 live feeds available for about two months from current date.

Periodically (monthly) exporting this M1 data to a CSV file from the history center will give you a good collection of reliable M1 data set over a period of time.

Back test conducted with the available broker feed should provide a good reference point to compare trades with live accounts. This kind of test study is especially useful if you want to change broker and would like to see performance between brokers. In this case one would just need to conduct two month backtest with Broker A (using feeds obtained from broker) and repeat the same with Broker B. Ideally the over all performance should be similar (not identical) between the two tests. Compare the results to the live forward tests available. This will give a good idea about how the EA will perform on the broker of your choice.

Our development team has conducted the following tests to serve as a sample study on the above lines (two month period - December, 2012 to January, 2013):

Third Party Live Forward Test (FINFX)

Tick data backtest (Dukascopy source)

Back Test Alpari platform (Feed from broker)

Back Test IBFX platform (Feed from broker)

Back Test PepperStone platform (Feed from broker)

Ray Scalper : Live Forward Test on FINFX by BestForexRobots.net

Start capital as on 1st December, 2012 : $5600

Account updated with V1.9 on January 14th, 2013 (equity under updated version shown in green)

Trade History can be accessed at : http://www.mt4i.com/users/rayscalper

Ray Scalper V1.9: Dukas Copy Tick Data Back Test

Start capital : $5600 (Set to match live forward test account)

Ray Scalper V1.9: Broker Feed Back Test on Alpari US

Start capital : $5600 (Set to match live forward test account)

Ray Scalper V1.9: Broker Feed Back Test on IBFX

Start capital : $5600 (Set to match live forward test account)

Ray Scalper V1.9: Broker Feed Back Test on PepperStone

Start capital : $5600 (Set to match live forward test account)

Summary of Back Test Results

Account	Test Type	Number of trades	Won	Lost	Net Profit
FinFX	Live Forward	33	27	6	$555
Dukas Copy	Tick Data Back Test	34	29	5	$650
Alpari	M1 Back Test	33	28	5	$586
IBFX	M1 Back Test	33	28	5	$527
PepperStone	M1 Back Test	29	24	5	$284

From the above data we can clearly see that all the back test (tick data and M1 backtests with broker feeds) are in close agreement with forward test account. Performance on PepperStone is considerably lower with 5 trades less than other accounts. We have received feedback on PepperStone real accounts missing some trades. The PepperStone back tests are in line with the feedback received.

The above study also shows that the EA's performance will be reasonably similar on most brokerages. Members can conduct tests on the above lines on their MT4 Broker platform if they find the live trades deviating beyond what is seen above.

Requotes, slippage and execution delays may result in some rare/occasional variation between accounts - This is broker related issue. If frequent requotes or delayed entries are noted, we recommend shifting the broker account.

The pre-launch version (v1.0) always made entries on M15 bar open. This was a good feature for easier/more reliable back testing using normal metaquotes data. The entry signal at new M15 bar also enabled the EA to be less dependent on broker feeds/requotes.

The drawback to the v1.0 strategy were

Suitable for 1411, but performance on 4411 affected due to entry at less than ideal price
Trade triggered at new bar open price, but actual price could be different
Use of fixed gap of 5 pips for trailing SL. Lower reliability of long term tests
Requotes did not affect entry, since the EA entered trade at any price after new bar open

Ray Scalper V1.9 makes entry on best price level based on every tick. This makes the strategy more suitable for using parameters 4411, which has better risk/reward ratio, while maintaining a good success rate of over 75% (even in 12 year back tests).

In v1.9, the EA will enter a trade only when price is suitable. In case of requotes, the EA will enter only until the price is within a suitable range. In some cases, depending on price action, the trade may not trigger in accounts with requotes because the price no longer is suitable for the trade signal. We may see some missing trades on such occasions. While this is a drawback, it makes the overall strategy better and more reliable. If frequent requotes, missing trades are seen, it would be better to use a different broker.

Ray Scalper V1.9 is a scalping strategy. But it is different from a typical scalper - the idea is to minimize broker independence by trading M15 time frame, having larger TP/SL levels and entry signals based on strong technical levels.

Ray Scalper V1.9 was released to members on 14th January, 2013. We have just about 3 weeks of trading (than 15 trades) to compare performance between different brokers and even with version 1.0. However, the results of live trades for the period between 14th January to 2nd February are presented below. Please note that the period of study (and number of trades) under consideration is not suitable for any analysis. This section will be updated after a period of two months.

Ray Scalper V1.0 (4411): Live Forward Test on FXOPEN (14th January to 2nd February)

Ray Scalper V1.9: Live Forward Test on FXOPEN (14th January to 2nd February)

Ray Scalper V1.9: Live Forward Test on Alpari US (14th January to 2nd February)

Disclaimer

U.S. Government Required Disclaimer – Trading foreign exchange on margin carries a high level of risk, and may not be suitable for all investors. The high degree of leverage can work against you as well as for you. Before deciding to invest in foreign exchange you should carefully consider your investment objectives, level of experience, and risk appetite. The possibility exists that you could sustain a loss of some or all of your initial investment and therefore you should not invest money that you cannot afford to lose. You should be aware of all the risks associated with foreign exchange trading, and seek advice from an independent financial advisor if you have any doubts. Clearly understand this: Information contained within this course is not an invitation to trade any specific investments. Trading requires risking money in pursuit of future gain. That is your decision. Do not risk any money you cannot afford to lose. This document does not take into account your own individual financial and personal circumstances. It is intended for educational purposes only and NOT as individual investment advice. Do not act on this without advice from your investment professional, who will verify what is suitable for your particular needs & circumstances. Failure to seek detailed professional personally tailored advice prior to acting could lead to you acting contrary to your own best interests & could lead to losses of capital.

*CFTC RULE 4.41 – HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT REPRESENT ACTUAL TRADING. ALSO, SINCE THE TRADES HAVE NOT BEEN EXECUTED, THE RESULTS MAY HAVE UNDER-OR-OVER COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED TRADING PROGRAMS IN GENERAL ARE ALSO SUBJECT TO THE FACT THAT THEY ARE DESIGNED WITH THE BENEFIT OF HINDSIGHT. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFIT OR LOSSES SIMILAR TO THOSE SHOWN.

By using PhiBase Synergy, you acknowledge that you are familiar with these risks and that you are solely responsible for the outcomes of your decisions. We accept no liability whatsoever for any direct or consequential loss arising from the use of this product. It's to be noted carefully in this respect, that past results are not necessarily indicative of future performance.