Ciao a Tutti,

today I’m showing the result of an **edge ratio analysis** I performed on the “PotenzaFX Groups” indicator. I’m showing the results of this analysis in the charts below (the vertical line is number of pips, but please note these are not equity curves) and then I will explain what this analysis means and how it was performed.

You can consider the** edge ratio analysis** as one of the “Ideas For Winning Strategies“. As traders we often look to new tools and indicators for technical analysis and we are prone to consider them as holy grails for out trading business. Indeed our eyes often tend to give more value to the big winning trades signalled by indicators and at the same time to overlook the importance of many losing trades. Just think for a moment to the most classic and simplest trading system (the crossover of moving averages) and you’ll understand what I mean.

The point is that we may spend time implementing a trading strategy based on a new indicator while we are not sure at all that this indicator really gives an edge over the market. Doing so in most cases we are simply wasting our time.

We’ve already shown some ideas about how to use the “PotenzaFX Groups” indicator effectively. I refer to the following two posts:

In this post I’m not suggesting a new trading strategy, but only showing a **statistical proof that “Groups” has indeed an edge over the market**. The edge ratio analysis consists in evaluating how much price moves in favor of a trade (**MFE = maximum favorable excursion**) and how much it moves against the trade (**MAE = maximum adverse excursion**) when the indicator signals a new entry.

Since Groups’ values go from -100 to +100, I’ve set -20 and +20 (a threshold that is 10% of the full range) as values for deciding short and long entries. When a candle closes with Groups <= -20 a short trade is opened at the beginning of the next candle. When a candle closes with Groups >= +20 a long trade is opened at the beginning of the next candle. After a trade is opened we measure MFE and MAE (as number of pips) until the next trade in the opposite direction is signalled.

I performed the edge ratio analysis on daily charts during 2011. A the beginning of 2011 we had the following groups of currencies:

- Group 1: AUD, EUR, GBP, NZD
- Group 2: JPY, USD
- Group 3: CAD, CHF

The Groups indicator is effective in trading Group 1 versus Group 2. Therefore I performed the analysis on 8 currency pairs: AUDUSD, AUDJPY, EURUSD, EURJPY, GBPUSD, GBPJPY, NZDUSD, and ZNDJPY.

**For each of the 8 pairs I’ve analyzed the cumulative sum of (MFE-MAE) during the whole year 2011**. Of course the bigger this sum the better, because it means that after a trade is signalled price tends to move in our favor more than it goes against us. **Each pair showed a sensibly positive sum at the end of the year** (about 3 thousand pips on average). The charts at the beginning of this post show exactly the cumulative sum of (MFE-MAE) for each pair. You can notice the pairs involving JPY have the best potential, while EURUSD and GBPUSD are the most difficult pairs to trade with Groups. Given these results we can say that Groups is a good indicator for building a winning trading strategy.

The chart below shows a sum of the edge ratio analysis across all pairs (the vertical line is number of pips):

We are currently testing an EA that exploits the “PotenzaFX Groups” indicator. The basic strategy works with a DLS (Dynamic Lot Sizing) approach, the same way as the “PotenzaFX DLS Trader” does. Another strategy uses fixed lot size, SL, and TP. We may add other strategies into this EA while we progress in our analysis of the Groups indicator and receive suggestions from our fellow traders 😉 I don’t know when this EA will be ready for a public release, but I believe it will not take much time. Stay tuned 😉

Ciao Andrea, Paolo

really interesting analysis, I’m testing right now the PotenzaFX DLS Trader in tandem with the “Group” indicator on Daily TF: when the spread between G1 and G2 is positive I set the Dls Trader only long for G1 and only short for G2, when thew spread reverse, the opposite; basically your 2Pfx strategy. I’ll do the test with different settings for the correlation filter.

When I’ll have a sizable period of analysis I’ll share my result.

Again let me express my satisfaction in regard of your products, really a step forward!

Enrico

Hi Enrico, thanks for your appreciation. I’ll be happy to know your results

Hi Andrea,

I’ve read this article with great interest and i’d like to submit a way to optimize your edge ratio analysis : Why not use Average True Range to get rid of the volatility bias as there is one accross all the pairs you mentionned ?

A way of doing this is by taking the daily value of every MAE or MFE and divide them individually by the value of the ATR at the entry point of the trade.

You then sum up every values obtained for each group by the number of days the trade was taken and finally divide the MFE volatility adjusted value by the MAE one to get a ratio that will indicate the potential gain we can expect to obtain for each trade.

For example, if it’s 1.2 for EUR/USD on a daily TF, we know we would have an average gain of 20% using your system with this pair and TF.

By the way, can you tell us what is the value of the y-axis on your last graph refering to because i have a hard time figuring it out…

Thanks for all you do.

Symphoenix

Hi Symphoenix, thanks for your feedback. The value on the y-axis of the last chart is number of pips, exactly like the previous charts. Indeed the last chart is just a sum of the 8 previous charts. Regarding the ATR, what you said is completely right. I must confess I used the ATR in my own excel sheets, but then I preferred to keep the analysis simpler and to cite only “number of pips” in the article. Of course if you use the ATR you can also figure out more easily an effective exit strategy, and so you can go from the edge-ratio analysis to a very basic (but complete) trading strategy. As an example if you find out that MAE < 2*ATR for most of the trades, while MFE > 3*ATR for most of the trades, then obviously a trading strategy with SL=2*ATR and TP=3*ATR may be a winning strategy. You must perform a backtest to make sure this is really the case. Good luck with your trading 😉

Sorry Paolo, i wrongly assumed Andrea was the author

Thank you for answering me back so honestly.

I was indeed wondering if the combination of PotenzaFX Groups for enters and another (like an ATR-based) one for exits would be profitable and maybe i’ll back test it.

I’m at the very beginning of my trading journey and i’m focusing right now on analyzing price action and trying to find an edge by the use of 4 different moving averages on a 15M chart. For now, my results are an average growth of 17% of my acccount each week and even if this is not so much, this is helping me to develop the right state of mind needed to be successful.

Of course, i bought all your indicators (except Paniere FX and the PotenzaFX EA) because i like the way you behave toward the community and wanted to witness my gratitude.

By the way, could you send me by mail your Excel sheet using ATR or is this too much to ask 😉

Anyway, keep up the good work.

Symphoenix

Hi Symphoenix, no worries. You cannot backtest our EAs based on PotenzaFX indicators because MT4 does not support backtesting with historical data coming from multiple currency pairs. We use other platforms for backtesting. If you make 17% per week that’s terrific, maybe you mean 1.7% per week? I’ve sent the ATR excel sheet to your email address 😉 Ciao, Paolo

Andrea/Paolo,

Comment on ‘Analyzng the edge of PotenzaFx Groups’

Extremely interesting – If I read these charts correctly it would seem to pay to avoid trading from the beginning of May to the end of August, because during that period the cumulative curve seldom rises and in general turns negative.

Am I correct or is that too simplistic?

Jim

Hi Jim, you can infer this kind of information from the excel sheet more than from the charts (there is a link to this sheet within the post). During 2011 the most difficult period seems June-July-August for most pairs. However there are also exceptions: some JPY pars reported a good (MFE-MAE) also during August. Anyway, I would not talk about any seasonality. We need to analyze at least 2-3 years of data before we can infer some sort of seasonality. Ciao, Paolo