Here we go! A newbie trade and journey journal

At the end of the day, trading is about pattern recognition. Trading education is about understanding which patterns matter and which patterns don’t.

Although we like to act like a good trading system can be broken down into hard rules and fed into automating software, this just isn’t true. If it were, there would be more programs for sale that genuinely make a profit.

I did an experiment that really drove this home for me. I had been back-testing a 5/8 EMA crossover system on the D1 chart. I had seen multi-year threads on other forums of people making a profit on this system, and yet my back-testing didn’t show that profit. So I went back and looked a bunch of the failed setups, noted the date, and then went in search of the financial news for that day. In many, many cases, there were clear signals (such as major scheduled news releases) that made it obvious to any discretionary trader that signals generated on that day would be suspect. But there was nothing on the charts that would tell you that.

Take out those failed trades and, voila, you had a profitable system.

In my view - and I don’t hold myself out as a master trader by any means, so take all this with a grain of salt - any attempt to run a completely non-discretionary system is just turning yourself into a bad, slow, stupid algorithm. You can’t out algo the algos. They have a team of geniuses constantly tweaking them, and they operate at speeds we can’t properly conceive of. Don’t try to beat them at their own game.

Our ‘edge’ is human discretion. It is drawing in all those ill-defined intangibles and knowing when to say ‘no’ when all the rules are saying ‘yes’.

What does all this have to do with your question? Well, there is a scene I remember from the movie ‘City of Angels’ - a bad movie, don’t see it - where Nicholas Cage asks Meg Ryan what the peach tastes like. She says to him ‘You don’t know what a peach tastes like?’ He responds, ‘I don’t know what a peach tastes like to you.’ Our brains are all wired differently. We see different patterns, even when we are looking at the same charts. You must, must, must find the patterns that work for you.

Some guys are all into support and resistance, or Fibonacci levels. I tried that, and a bunch of other things that different people have sworn by, and it just didn’t work for me. It works for them, but not for me, because what they see and what I see are two different things. And I suspect that if a lot of those guys tried what seems to be working for me, it wouldn’t work very well for them.

So, yeah, you need to find your own patterns. That doesn’t mean they have to be unique - they have almost certainly been found and used and talked about by other people - but they do need to fit you, and the way your brain works.

2 Likes

So basically I need to find my own patterns, but they can be the same patterns other people use. So a system based on entering on pullbacks and using 1 ATR stops and 127% Fibonacci extensions can potentially be profitable.

Based on your post, I can extrapolate this process.

  1. Create an idea for a trading system
  2. Backtest on a reasonable amount of data (say 100 trades)
  3. Examine failed trades for possible causes (such as news releases)
  4. Tweak system to reduce failed trades
  5. Retest

So to be a profitable trader, I don’t need to find the next animalistic harmonic pattern or develop some indicator or some other in-depth work like that. I just need to repeatedly test ideas, keep the best ones, and throw away the rest.

Broadly speaking, yes. I’d tweak those points a little, though.

  1. Steal an idea for a trading system has reputable adherents and a proven track record. Some ideas are just bad ideas, so filter those out by focusing on the ideas that seem to have worked for at least some other people.

  2. Back-test for a reasonable period, as relevant to your chosen time-frame. Conditions change the further we go back in time, so going too far back doesn’t actually tell you a lot more about how the strategy will perform today.

  3. Examine failed trades for red flags. You don’t necessarily need to understand why things went wrong, but you would like to identify factors that could potentially warn you when a future trading signal might be misleading. You are particularly interested in factors that are common to a significant number of the failed trades.

  4. Consider the performance of the system if a realistic number of the failed trades are removed from the equation. If profitable, consider forward testing. I think back-testing is useful as a basic ‘sanity check’, but I find forward testing far, far more useful for actually refining a system.

3 Likes

Post of the Year!

The Ever Giving Kudos VIPER

I have off from work tomorrow through Sunday, so my goal is to get 15 hours of backtesting in. The initial idea I am going to test is going to revolve around getting in on pullbacks. Pullbacks stick out for me like a sore thumb, so that is as good as any of a place to start with.

My idea is that if I can figure out the best time to enter the pullback then I can ride that move up to profit targets. I definitely want to set stops and targets because I can’t watch my trades all day long. So my question is how to determine when a move against the trend is really a pullback and not a reversal?

My testing is going to test a lot of different things at once. For entry signals, I plan to test double bottoms/tops, RSI, Stochastic, CCI, Fibonacci retracement levels, previous support/resistance levels and single candlestick patterns. Each thing I look at is going to have the goal of helping me fine tune the entry. I even plan to include whether or not the direction of the trade corresponds to a higher fractal trend. For targets, I plan to test ATR based targets, Fibonacci extensions, harmonic moves (where second leg is equal in length to first leg), previous support/resistance levels, and round numbers (such as 1.2600 or 1.1700). For stops, I plan to test ATR based stops, Fibonacci inversions, and previous structure levels.

Once I get a good amount of data I should be able to see which combination of factors gives me the best risk/reward ratio or highest win rate (it would be really nice to get a 5:1 ratio with 85% win rate :slight_smile: ). From there I can develop and refine my trade rules to test in my demo account before trading live again.

The good thing about all of this backtesting is that it should carry me through the screwy holiday markets and back to a more normal market behavior after New Year’s :slight_smile: Also, I hope it will make me a lot more competent at executing trades than I am now so that I can focus on risk control and money management at that point.

1 Like

Excellent one. Are you using basic MT4 functionality for backtesting your strategies? Is there any better software or platform?

Analysis paralysis sucks… that is all.

I do use MT4, but I back-test manually. I want to see every single trade.

Who knows what you might learn if you actually look at stuff. :smiley:

It could be worse. I watched the GBPJPY get 90% of the way to my profit target, only to turn around and head the other way. I’d moved my stop to break-even, so it wasn’t a loss, but I was kicking myself for not getting out as soon as the reversal started.

No matter how many times I tell myself ‘when in doubt, get out; hope is not a strategy’, it never seems to sink in as well as I would like it to!

Well, hindsight is 20:20, as they say.

That does suck.

Right now I am trying to formulate a strategy for trading pullbacks using CCI and Stochastic indicators with Fibonacci levels and support/resistance lines. I pop both those on my chart and I am looking at my first trade setup. I spot it, and now I am trying to capture the data in a spreadsheet to analyze.

That’s where I am getting stuck at - what data do I capture, and what can I ignore? Right now my spreadsheet as almost 20 columns of data and I have no idea if I am even getting the right stuff.

And now my brain is telling me to go grab a free system from the forums here and try that instead… grr…

After shaking off the flu over the weekend, I decided that it’s time for me to take a break from trading real money. My strategies are untested and my execution needs more practice. At this point, I am trying to run before learning to walk. My plan at this point is to sit down and spend some time finding a few different strategies that really work for me.

By that, I mean strategies that I feel good about, that I can readily identify in the market, and that have a track record of being successful as measured by back testing and demo trading. Since the market has only three states - bullish trend, bearish trend, consolidation - it stands to reason that I have one or two different range strategies and a strategy for trading the trend.

My goal is to have all this done by 3/31/18 and resume live trading at that point. This week, the goal is to re-read all of the school section here and look for techniques that I want to investigate. Next week, the goal will be to backtest a trend strategy and the week after that the goal will be to backtest a range strategy. Once I backtest those strategies then I will spend a few months demo trading those strategies, providing that backtesting shows a positive expectancy.

Sorry to hear about the flu. A couple of suggestions, though.

First, don’t set arbitrary time-lines for yourself. I know that I’m being hypocritical here, because I fall into that trap myself all the time, but your strategy will be ready when it’s ready. Feeling like you have missed your self-imposed deadline is just a way of beating yourself up.

Second, I don’t think it’s necessary to have a strategy to for all market conditions. Better to take your winner and look for markets that fit your criteria (or simply sit out a market that does not suit your edge) than to use second-rate strategies simply because you want to be ‘in’ your chosen markets.

That last one was really tough for me to wrap my head around, but every day I sit out is a day I don’t lose money.

Hey Bud, the Flu is not good, glad you are on the mend. A suggestion, since we are in a range market, especially the EURO, maybe you might want to look at that first since it is happening right now. But, however you work best, is the most logical way of approaching this.

The Ever Suggesting VIPER

1 Like

The thing is I don’t know how I work best. The early Forex YouTubers I watch always advocate for hours and hours of backtesting, followed by demo testing, followed by live trading, but that just might be for liability purposes. I suppose live trading small positions and controlling risk would give me the same practice and potentially make me some pennies.

I really dislike backtesting because its tedious, but if it is worth it, then I probably should do it. I am just not sold on backtesting being worth the time.

Wow, it’s been a few days since I updated this thread. I spent last night watching the market to see how a new pattern I learned works. Very informative, and very helpful.

I am also thinking of making another YouTube video where I just ramble on into the mic about how I am looking at the market. I might do a live stream as my skills progress and I feel comfortable teaching other people how to trade (for free - I want to make my money trading, not selling failed systems).

I also talked with a friend that I met here about prop firms and got confirmation on something I longed believed - prop firms are brutal places to learn to trade. However, I see the benefit of talking with other traders IN REAL TIME while trading and by doing so we can all learn from each other. To that end, I will gladly setup a Discord voice chat server if anyone wants to hang out with me while I trade.

My standard trading hours are 1800 to 2100 (UTC -6) on Mondays, Tuesdays, Wednesdays, and Thursdays. Of course, these are not set in stone and will vary - without warning I might add. If you think you would be interested in a trader hangout type of room (and you alone are responsible for losses if you follow anyone’s trade) PM me.

That’s midnight to 3 am GMT ? - That’s Asia Hours ! - Are you looking to trade that interval, or discuss pending entries and exits on larger time frames ?

I’d quite like to play - but am on GMT and by midnight - I’ll either be Pssst or asleep !

Best to all
F

Yep, that’s the only hours I can trade given my work schedule. I might do the odd Friday or Sunday afternoon (1600 to 1900 UTC - 6) once in a while, but not reliably. Of course, if enough people want a Discord server to hang out on it doesn’t matter if I am there or not.

Productive night tonight. So far, banked 10.9 pips on the AUDUSD pair, and looking at one more potential trade setting up right now. Fingers crossed that I don’t give those 10 pips back :slight_smile:

I haven’t figured out yet how to export chart images from fxTrade platform yet. I might yet be forced into MT 4. And I got in on a double top play for the last trade of the night. The goal of this trade is 10 pips with a 5 pip stop loss.

The first trade was a short entry at 0.7587 (19:33 UTC -5) for 1,000 units. Closed 500 units at 0.7575 and closed other 500 units at 0.7578. First 500 units net 12 pips, and second 500 units net 9 pips. Overall profit 10.5 pips (assuming averaging the two positions together is proper). Overall profit in USD is $0.78 cents.

The second trade was a long entry at 0.7581 at 20:14. Closed entire position at 0.7584. Net profit of 3 pips. So far up 13.5 pips for the night. Net USD profit on this trade was $0.31 for a total of $1.09 overall.

Double top trade entered short at 0.7586 (20:30). This trade was a test of nerves as price came up to a hair away from my stop loss, stopped, and then started moving back towards break even. Now with 8 minutes left in my trading session, this trade is at -$0.37. Right now I will be happy with break even within the next 8 minutes. This trade hasn’t been in the green since I put the position on and in 7 minutes that will be two whole bars of the trade moving against me. On these short trades, I like to see at least a move into profit within the first three bars.

And the trade is stopped out 2:01 minutes before the buzzer for a loss of 6 pips, or -$0.63. Net profit for the night is actually a profit! Net profit of $0.46.

So yes, it turns out I can make money in the markets. I got a long way to go yet, but this is definitely a positive step in the right direction.

1 Like

I snuck in a trade today at work. I wanted to experiment with higher time frame charts to see if that would allow me to take advantage of some of the market volatility from the New York and London sessions without actually being present during those sessions because of corporate slavery and all.

I found a range trade on USD/CAD 4 hour chart. I entered short at 1.29002 (top of range, and psychological number) at 12:32 PM (UTC -6). My stop is at 1.29280 and my profit target is at 1.27000 (200 pip move if I catch it all) but I have alerts set at each structure level on the way down to check for loss of momentum in the move. This trade is currently 6.7 pips in profit.

1 Like

Good lad :slight_smile:

It’s in the nature of these things that we keep trying something different. The more we trade - the more we get accustomed to trading.

[edit - after a while, we learn the types of trade and timescale which work best for our own personality. ]

1 Like