Journal - Crossover system

Trading Discussion Trade Journals
1

A disclaimer, before I begin.

I am not a trader. I’ve been playing on a demo account for about two months, which I’ve opened right after finishing up BabyPips’ school, Forex4Noobs free course and after watching several webinars on YouTube. I’ve also blown my demo account. What this means is that I don’t want to get anyone’s hopes up: there is a good chance that everything I’ll do here is going to be wrong, and that my setups will result in net losses.

With that being out of the way, if you want to follow the wacky journey of a random 23 years old Italian guy trying to learn how to trade the markets, keep reading. I’ll be posting my trades as I open them, keeping track of my profit as time goes on. But first, I’ll give you a rundown of the system I use. It is a small variant of the “Amazing Crossover System” by Forex Phantom, so credits to him.

Account.

Demo account on Oanda. 500 Euros starting capital, 1:30 leverage. MetaTrade 4 as my platform of choice.

##Money management.

I will risk 2.5% of my equity on each trade, which is 12,5 Euros. I will enter trades with at least 1.5 Risk to reward ratio. For each 50 Euros gained or lost, I will compound my equity to adjust for risk. This is mainly to keep things simple, and to give myself milestones to look forward to.

##Pairs.

I will trade EUR/USD, USD/JPY, GBP/USD and USD/CHF exclusively.

##Timeframe.

I will trade the 1 hour chart. I will use the 4 hour chart to assess the overall trend of the market.

##System overview.

It is a simple strategy based on the crossover between two exponential moving averages on the one hour chart. It uses the RSI and MACD as signal filters, as well as the ADR to determine stop losses.

##Indicators.

Exponential Moving Average - 5 Periods - Applied to Close - Yellow. It will be referred to as Fast moving average.

Exponential Moving Average - 10 Periods - Applied to Close - Blue. It will be referred to as Slow moving average.

Relative Strength Index - 10 Periods - Applied to Median Price (HL/2). No overbought nor oversold levels. Only one level, placed at 50%.

Moving Average Convergence Divergence - Fast 5; Slow 9; Macd Sma 4 - Applied to Close.

Average True Range - 12 Periods.

##Long signal.

Four conditions have to be met.

  1. Price is making higher highs and higher lows on the 4 hour chart.
  2. Fast crosses Slow from below on the 1 hour chart.
  3. RSI is approaching the 50% level from below.
  4. MACD shows confluence with the RSI.

Entry is made when RSI crosses the 50% and MACD shows confluence.

##Short signal.

Four conditions have to be met.

  1. Price is making lower highs and lower lows on the 4 hour chart.
  2. Fast crosses Slow from above on the 1 hour chart.
  3. RSI is approaching the 50% level from above.
  4. MACD shows confluence with the RSI.

Entry is made when RSI crosses the 50% and MACD shows confluence.

##“Don’t trade” signal.

I won’t take a trade if MACD, RSI and the EMAs are flat. As in, not showing a lot of movement. This suggests that the market is in consolidation, and I won’t enter a ranging market.

I won’t trade during the release of Interest Rate decisions and Non Farm Payrolls. I also won’t trade on Fridays after 18:00 GMT +1.

##Stop.

ATR * 1.5. 20 pips minimum. I will manually trail my stop to breakeven.

##Target.

Stop * 1.5, at least. Target is placed a few pips before the closest support and resistance zone.

3 Likes
2

Interesting - good luck with it!

Just a couple of questions, if I may?

  1. You haven’t backtested this, I think?

  2. Why, if your fast and slow MA’s are 5 and 10 periods, are you using 5-9-4 for the MACD settings? Does that add anything or just duplicate what you can see from the MA’s? Would different MACD settings from the MA settings (like perhaps 20-40-30 or 15-30-25) instead perhaps add something, and “qualify” the signals by independent corroboration? I’m not saying that would be a profitable system with a genuine edge (and honestly I don’t think it would be) but at least that way, the MACD would be an additional filter, wouldn’t it, rather than a duplication?

1 Like
3

No, I haven’t. Robopips has backtested the original version, which is pretty much the same without the MACD and the ADR. You can check out his results here.

To be brutally honest, I simply took the settings of the original system and went with them. However, why do you say that the MACD would replicate the EMAs? The lines on the MACD are not moving averages of price. What am I missing?

4

A filter.

The MACD lines aren’t MA’s of the price, but they’re dealing with the same MA’s that you’re already using. (Disregarding the difference between 9 and 10 periods)

The MACD line is the difference between 5-period and 9-period MA’s (you already have this on your chart), and the signal line is just a 4-period MA of the MACD line. It isn’t a filter.

MACD settings of 20-40-30 (say) would at least be a filter. (But still wouldn’t confer a genuine edge, to be honest.)

I don’t want to come across and hostile or critical, in your Journal, so I’d better not comment.

Except perhaps to extol the joys of BACKTESTING.

It really, really is worth learning how to do that, Giovanni.

But good luck, anyway.

1 Like
5

What would you suggest as a good filter?

By all means.

As I’ve said, others have already backtested the system. Besides, I’m testing this on a demo account, it’s not like I’m going to lose money. I will probably backtest when the markets are close, using the replay function on TradingView.

6

Personally, I think you have selected a good basis for your trial. I may be wrong but I assume, at your stage of learning, that you are more interested in gaining an understanding of the dynamics of trading forex rather than simply looking to “make money”.

From that perspective, I think your approach is good because it is simple, contains a variety of commonly used indicators (but not too many), and on a timeframe that will give you sufficient signals without an overload of fake outs. It also brings in a good habit of considering higher timeframe conditions.

You have also included risk/money management parameters which will give your practical experience how these function.

This little method should instill patience and discipline as well as the good practice of recording your trades and analysing the results.

This are all very good ingredients on an instructive demo account.

The only things that I think you may like to look at as time goes by are the use of support and resistance lines in fine-tuning your stop and target levels and whether a different approach to just 1.5 x stop distance for your target might be more optimal - but what you are suggesting is fine for starters, I think (being a demo account).

I think the basis you have outlined is a good starting point and a very good way to approach demo trading. As time passes you can gradually see what works and what doesn’t and, most important, why -
and develop it stage by stage.

I also think your English is brilliant if it is not your native language as an Italian? But that doesn’t affect your system so very much! :smiley:

Good luck, I will be following your progress with great interest! :slight_smile:

1 Like
7

Yes, that’s right. I know I won’t be able to trade with a decently sized capital for at least a couple of years, since the only source of income comes from the job I do at my uncle’s and that’s pocket change. All I want is learn how to trade so that, after graduation and after I get a stable income, I will know what I’m doing.

Now that I think about it, yes, I could plot a single zone towards my target and use it as fixed take profit.

Thanks a lot!

Also, do you agree with LukasVisser about the MACD? I’m worried I should change its setting or get rid of it.

8

One or the other, maybe? I think it makes little sense for the moving averages from which it’s calculated to be the same ones you’re using in their own right? (Lukas’s suggestion of 20-40-30 sounded interesting to me? And would be a “filter”, if that’s what you want, in that it would illustrate a different scale of time?)

1 Like
9

I’ve tried 20-40-30, and the MACD becomes a bit weird. It stays permanently below the center line and filters out pretty much all the signals.

About my current settings, I’ve noticed that the MACD line often disagrees with the EMA crossings. Not by a lot, sure, but enough to filter out RSI crosses.

[Album] Imgur
10

If you think about it, I think you’ll agree that in the long-term, with any currency pair, it has to be above/below the midline on approximately a 50/50 basis, doesn’t it? You’ll need to look at a minimum of 200 entrie,s before making any conclusions approaching statistical validity.

With a stock index, it would arguably be above the midline more than 50% of the time, on some timeframes, since the indices rise more than they fall, but clearly that isn’t the case with currency pairs?

Anyway, probably time for me to bow out of your thread, leave you in peace, and just wish you good luck with it.

1 Like
11

That would be ideal if I was swing trading. However, when looking at the 1h charts with the settings you suggest, I would get something like a signal per month. I would have to change the RSI aswell at that point, because you’d have two indicators who are analyzing two vastly different time ranges. At that point, however, I am no longer trading the same strategy and should re-evaluate everything from the ground up, not just changing the indicator.

Thank you, but I do appreciate any criticism. I’m here to learn, not to have people tell me how good I am!

1 Like
12

Ahaa! This is already your first deviation from the trading education “straight and narrow”!!! :smiley:

The best person to (eventually) answer this question is actually…YOU! Through your own research! That is the whole point of this kind of demo exercise.

There are contributors on this site like @LukasVisser, @LaughingCharlie, @tommor, @_bob, and others who have, collectively and individually, decades of experience of markets and trading. It is always wise to listen to their comments but you will not always get the same response to issues. And where there is a divergence who can say to you who is correct?

This is a typical situation where you take heed of the issue raised by a reputable commentator and investigate it. You should dig deep into the structure of RSI and MACD, see their similarities and differences and what they are really intended to show and then compare the settings you are using and how you are reading them, etc. But, ultimately, it will be through practical application and results analysis that you will determine what has added-value in your trading model.

As a rule, there is little point in using multiple indicators that are suggesting the same thing. All you will get is slight differences in signal timings which creates confusion over which one is “correct this time”. Obviously, if one indicator is consistently more correct than another then the answer becomes clear what to jettison as baggage.

Another point to bear in mind is that these indicators are precisely that - indicators! They indicate i.e. suggest, that something might be happening that you want to know about. But you should still use your key indicator to evaluate what it is suggesting - your brain. You know that you are using very short term MAs on a short time frame in markets that do not move like perfect sine waves and therefore there WILL be fakes whatever filters you use.

In the context of what you are doing here and that you are on Demo, if I were in your position now, I think I would put Lukas’ comment behind my ear as something important to evaluate as trades evolve. But because I am starting with a core method put together and already supposedly backtested by someone else then I would start off keeping to all the parameters contained in that method. Then you can systematically evaluate all the individual parameters in your forward testing as your progress. And whenever you ultimately reject, tweak or enhance things, you will know why…

PS: I don’t use MT4 myself but I vaguely recall reading somewhere that the standard MACD on MT4 does not include the signal line? This is an important component of the general MACD concept but I don’t know if it is used in the Crossover Method that your model is based on - but might be worth checking…

2 Likes
15

Day 1 Recap. (-1.4%)

Starting balance: 500.00 €.

Closing balance: 492.94 €.

Equity adjusted for open investments: 498.65 €.

Profit: -7.06 €.

Closed Trades

1. USDJPY - Buy. Result: loss, - 5.92 €.

2. USDCHF - Buy. Result: loss, - 5.81 €.

3. USDCHF - Sell. Result: win, + 4.42 €.

4. USDJPY - Sell. Result: loss, - 3.52 €.

5. USDJPY - Sell. Result: win, +3.95 €.

Open Trades.

6. EURUSD - Buy. Currently at + 5.45 €.
Stop trailed to + 3.06 €.

Winrate: 50% (Sample: 6 Trades)

Notes

I’m not content with today’s results. I think I should be more discretionary with my trading, regardless of how mechanical the system is. I took too many trades for a single day, and that can’t be a good sign. I’m most likely too blindly following the signals given without taking market conditions into consideration. I’m hopeful for tomorrow!

Also, note to myself. I do have to gain a deep understanding of all the elements that compose the system I’m using: the RSI, the MACD, the EMAs and the ADR, as well as price action. I feel like what I know is not enough. I want to get to the point where I possess the knowledge to tweak their settings on my own judgement.

16

[quote=“giovannicali, post:15, topic:144365”]
I’m not content with today’s results. I think I should be more discretionary with my trading, regardless of how mechanical the system is. I took too many trades for a single day, and that can’t be a good sign. [/quote]
I wouldn’t be too hard on yourself about yesterday. It was typically one of those days when there was very little movement, a very small range, and therefore the signals were generating in an aimless market - this is precisely what is the Achilles heel of this kind of method. You cannot know beforehand whether a new signal is the start of a new move or just simply because two MAs are always on one side or the other of each other, simply by definition!..at least you have seen how effectively these kinds of days, that actually go nowhere, can significantly eat into one’s profits!!!

But I think you can console yourself that at least the USDJPY is currently in a prolonged state of consolidation which is typically characterised by directionless volatility and whipping. You can see this on this daily chart since the start of the year. Yesterday’s range is arrowed in blue. The blue and red lines are the daily 5 and 10 ema’s which show how this pair has been in a prolonged and uninterrupted downtrend since January. But there was a large candle on 28.3. which ended that and pushed the pair into a range consolidation that has been continuing ever since. These sorts of long trends rarely reverse instantly on the spot and usually go through a period of uncertainty that can drag on for a long time. You can see this here where the price range is embedded in that pink cloud. There have not been any signs of significant profit-taking in the price action yet and this could just as easily continue down as reverse and move up - hence the indecisive intraday price action.

UJ 09042018.png1386×878 16.7 KB

[quote=“giovannicali, post:15, topic:144365”]
… I want to get to the point where I possess the knowledge to tweak their settings on my own judgement.[/quote]
Just one word of warning here. With this kind of method it is only too easy to look at the short term results (in this case only one day so far) and assume it is the method that is at fault and start to tweak the indicators to “correct” them- but what you are actually doing is just curve-fitting, where you keep adjusting the indicators so that they would have worked “yesterday” - and the end result is that they now do not work “tomorrow”…

Market characteristics do change over time and it is worth keeping one eye on how the results are coming, but this should not become an ongoing, short-term habit and normally may only need attention after prolonged periods or seasonal changes, etc.

On the other hand, it is naturally a good thing to always look at the market and decide when are the optimum times to apply the method, such as this monitoring of the daily price activity, but that is a different thing to tweaking the system itself.

I hope you don’t mind me intruding into your journal…

1 Like
17

I wonder how I should react to this. Should my reaction be “I have to find a way to better filter the system”, or “I should scrap the system because of how poorly it performes in a ranging market”?.

That’s not what I meant. As LukasVisser made me realize, I am using this system with standard settings only because that’s what the guy who created it came up with. I didn’t question the strategy, and my knowledge about the indicators is the one I have thanks to the school. I know how to use them, but I have no idea about how they work.

I feel like I am driving a car, and perhaps I am a good driver, but I know I wouldn’t be able to fix its engine should the car stop in the middle of the road. I believe that learning the ins and outs of the indicators used by this strategy would provide me with more confidence, as well as representing a nice educational experience.

Plus, I’m already decently educated on price action trading, meaning that I can analyze each candle with logic. My main source of education is Nick Bencino and his webinars, about which I love the idea of each candle being the story of price as opposed to being part of predermined patterns.

That turned the way I look at charts around. Recognizing reversal patterns is being able to drive a car; knowing how to reach each individual candle and the information about price it gives, means that I have a much deeper understanding of the subject.

Ha! Not at all, quite the contrary. I enjoy feedback, as I’m here to learn and be criticized.

1 Like
18

Don’t beat yourself up too much over that: it’s probably true of at least 90% of retail forex traders.

2 Likes
19

I’m just trying to evaluate my knowledge in an objective way, precisely because I don’t want to be like 90% of retail forex traders. :smiley:

2 Likes
20

Day 2 Recap. (+3.30%)

Starting balance: 492.94 €.

Closing balance: 509.99 €.

Equity adjusted for open investments: 509.99 €.

Profit: +17.95 €.

Closed Trades

6. EURUSD - Buy. Win: +2.99 €.

7. EURUSD - Buy. Win: +12.22 €.

8. USDCHF - Sell. Win: +5.38 €.

9. USDCHF - Buy. Loss: -3.13 €.

Open Trades.

None at the moment.

Winrate: 56% (Sample: 9 Trades)

Notes

I am waiting for a signal to form on GBPUSD (Imgur: The magic of the Internet). It has been two days since the last signal, which kickstarted a very nice bullish move I unfortunately missed. I am glad I resisted the temptation of getting in long after the signal has printed.

One thing that I’ve noticed about USDJPY, is that signals are much less reliable if they came shortly after another failed cross. This is suggesting me it might be a way for the system to tell me the market is unsure about where to go, and I’m considering adding it as an additional filter. Something along the lines of: entry only if the signal has not been preceded by a failed cross in the past 10 candles.

Another thing. I happened to stumble upon a video about the Volumes Indicator (which is actually based on Tick Data). It seems like this can be used to measure how easily a candle is formed, representing the struggle between supply and demand. That could be a very nice addition to my price action analysis, although I fear I might be filling my charts with indicators.

1 Like
21

Sincere congratulations! :slight_smile: What a difference one day can make! :slight_smile:

How are you finding it, watching 1 hr charts? Is it difficult and tiring? Are you selecting certain trading session times to watch for signals?

1 Like
22

It’s not as bad as I thought. Yes, I do have a lot of stuff to do everyday (work, college, my side business), but I do find that most of the signals I trade happen a few hours after I wake up - so that I can trade them right away in the morning. Other than that, I just check my smartphone once every two hours. If I find a signal, I open a position or just take notes. Since I’m only watching four pairs, and since my MT4 app has already been set up, I only need about ten seconds per chart. That’s less than a minute every two hours. It’s manageable.

I’m also developing a habit of splitting candles in half before entering a position. I look at M30 charts to see how the candle I’m basing my entry on was composed. That kept me out a couple of bad trades today - and for that I thank my price action knowledge!

Thank you!

1 Like