Dynamic Probability Trade Strategy

[expiated]

The lower chart is the setup I am using presently (after making modifications on Monday). Instead of using three different price range envelopes as I was doing initially (the top image has five envelopes, which is why it is so confusing) I am now using one envelope at three different deviation levels.

This initial setup (from Saturday?) has certain strengths that the one from Monday cannot match, so I believe that maximizing the clarity with which I am able to trade is going to demand that from now on I use both of the setups in tandem with one another.

 

[expiated]

I was disappointed earlier when it appeared that this would be a losing day, but it looks like everything is now back on track.

 

[expiated]

I have decided to add to my setup three proprietary moving averages I had laying around from some of the earlier coding I did in the past, which I think I'm going to call my "Clarity Lines."

BEFORE

AFTER

 

[expiated]

I have decided to add to my setup three proprietary moving averages I had laying around from some of the earlier coding I did in the past, which I think I'm going to call my "Clarity Lines."

So far, looking to the "Clarity Lines" to aid the decision-making process appears to hold promise. In fact, currently my average profit trade is almost double the size of my average loss trade, so I'm anticipating evaluating this new setup between now and March to determine how typical/consistent such results might be.

 

[expiated]

My charts don’t mesh with Cristain’s in this instance...



My hourly chart (above) suggests that CADJPY is in the upper region of the typical day range, so I would not be inclined to enter a long position.



Moreover, the clarity lines plotted on my other chart arrangement presently appear to be saying that the pair might be all through with climbing for a while.

 

[schweiz]

I don't get it ... You use your indicators but it's essentially a discretionary system? So cannot be tested? What's the point then? You may be an amazing trader with or without these indicators.

I don't agree that discretionary systems cannot be tested. You should only interpret the results differently.
Also the degree of discretion is important to know. 10% discretion is totally different from 90% discretion, as well as their impact on the results.

I am a 10-15% (estimation as I have no clue how to define the percentage) discretionary trader, as I do the last step manually.
I tested my system various times. Each time the mathematical part was exactly the same, but the final entry or exit could vary slightly. If you see that all the results are in line and the variation of the results are almost the same, you can assume that with high probability the results in realtime trading will be in line too. Especially because the big losses come from hitting stops. And stops are not discretionary, they are fixed before every trade starts.

These numbers vary slightly in the tests:

• number of winning trades
  
• average profit per winning trade
• average loss per losing trade
• trades per day

The variation is a result of small differences in execution prices, and also in taking or nothing taking some trades that are not clear on the discretionary part (which means overrule the system and skip the trade or doing a trade I should not have done). The total impact of the discretionary part is small as some decisions will make extra money while others will make extra losses. The extras on both sides will compensate mainly each other and flatten out the final impact on the results.
And if you take the results of the worst test you have a high probablity that your real results will be above these.

 

[expiated]

I don't agree that discretionary systems cannot be tested. You should only interpret the results differently.

Not to mention that if a trader tests a system that has discretionary aspects to it, there is the theoretical possibility that this could lead to the inclusion of additional rules which make the system less discretionary going forward and result in a subsequently higher success rate.

 

[expiated]

My charts don’t mesh with Cristain’s in this instance. My hourly chart (above) suggests that CADJPY is in the upper region of the typical day range, so I would not be inclined to enter a long position.

Cristian Moreno's daily call notwithstanding, CADJPY did indeed fall from yesterday's open, as suggested by the Dynamic Probability Trading forecast model...

At this time however I think it IS reasonable to expect that the pair will gradually climb higher in that the day-to-day trend has just turned bullish. (Also, seven hours ago the candlesticks bounced off the lower bands of the inner two price ranges.)

 

[expiated]

As an intraday trader, anything beyond a 60-minute chart is looking too far into the future for me to find the information actionable.

Yes, USDCAD is bullish in the long run, but my forecast model suggests the day-to-day trend just turned bearish...

Moreover, the lower bands of my inner three price ranges converge around 1.3172, so this is the area to which the rate would presently need to drop before I would consider entering a long position.

Technically, I could justify buying the asset if I get an upward hook in the yellow intraday trend line/moving average, but honestly, I'm just not all that crazy about the pair's structure as it currently stands since the rate is kind of sitting in the middle of everything right now.

 

[expiated]

I am very hopeful that the system will continue to work, so the main function of this thread will be to help me maximize and standardize the way in which I employ it.

This thread has served its function, so this might be its final entry.



The only loss I incurred over the last 24 hours was due to my putting on a trade as I was retiring for the night. Had that not been the situation, I might have avoided it, yet it was a valuable loss in that it provided me with the last bit of insight that enabled me to reach the degree and level of clarity I was hoping to achieve.

The configuration of my one-hour charts is now set in stone. (I’m actually using two almost identical versions interchangeably depending on whether I’m simply monitoring the charts for reversals or analyzing price action within the intraday trend.)

However, the “real” analysis occurs in the context of a five minute time frame. Unlike my hourly charts, which hold only two moving averages (along with three or four price range envelopes), the last bit of clarity provided by this morning’s insights came as a result of my plotting a grand total of TEN moving averages on my five-minute charts!

That’s going to sound crazy to a lot of experience, knowledgeable, and successful traders, but it is what it is.

The moving averages reveal what is going on with trends within trends. They let me know (along with two sets of simple moving average envelopes with deviation bands set at multiple, strategically defined levels) in which direction price is probably headed in both the short run and the long run, whether or not the market is likely being controlled by strong fundamental influences, when the variables are so extreme that it is only a matter of time before rates will inevitably be overcome by forces of regression, when volatility is lacking and pullbacks are likely to be significant or momentum is so strong that pullbacks are sure to be short-lived, etc., etc.

The five-minute setup combined with my one-hour charts tells the whole story, and for me, that IS the whole story. So there’s nothing left to do now but look for other distractions to pass the time.