• Buying pullbacks or breakouts

    During my recent holidays in Greece, I had some free time to continue writing the new book. This let me to improve the trade setting software that I had developed, in order to include a Risk/Return calculator.

    As you know, my best trades are to buy leading stocks that are pulling-back to the 5MA while on a positive EV pattern. This has worked well and what works should be kept.

    Within a few days, I wrote the necessary code to calculate not only the Risk/Return of an investment strategy but also the efficiency level of that strategy.

    The R/R ratio simply shows the ratio of your gains compared to your potential losses for the use of that strategy. The software adjusts the trade's holding period to maximize the R/R.

    The strategy efficiency level is the return that this strategy would have produced in the past compared to a simple Buy/Hold strategy, taking into account the invested time during that period.

    For example, if for the past 100 days, a B/H strategy would have producer a 10% return and if during that same period our investment strategy produced 5% but was only invested 20% of the time, the investment efficiency would be: (5% / 10%) divided by 20% = 2.5. This means that the investment strategy could have only caught half of the trend's profit, but since money was used only during 20% of the time, it was a strategy that was 2.5 more efficient than a B/H strategy.

    I then improved the software to include different strategies: PB (PullBack), BO (Breakout) and BOV (Breakout on volume.) The BOV strategy matches Mike's conditions used for the volume alert.

    Over the week-end, I took some time to test the software on different stocks and see what strategy works best for what type of stock. I will not really generalize my findings on the base of just a few examples, but what I suspect is that breakouts work much better than pullbacks on momentum stocks, but that mature large stocks are best traded on pullbacks.

    Let's see two examples: YY, PAA and MSFT.

    YY is a typical momentum stock. It is an IPO of 2013. As usual, we only trade long when the price is above the 50MA.



    We can see that on Thursday of last week, YY broke out from a base formation of a C&H pattern.



    The calculations below are based on the close of Wednesday (One day before the brea-out.) The ideal breakout price was $91.

    The Table below shows the results of trading a PullBack (PB) to the 5MA, using a 5% stop. We can see a good R/R but a poor strategy efficiency.



    Using a breakout above $91 - To simulate past breakout for the past 400 days, I of course calculate the breakout distance from the last close - we can see that there is no improvement in the results of both R:R and the Strategy Efficiency. Note that the probability of a BO is higher than 30%



    If we now evaluate teh returns of a BO on high volume, we can see that bot the R/R and the Strategy Efficiency more than doubled. The probability to have such high volume BO however fell to 12% (51 possibilities for the past 400 days.)



    Unfortunately, this is not the end of the story. Indeed, if we look at what happened on Thursday, we can see that $91 was soon reached after the opening and that at 10:11, we were already at $93. This means that trading a high volume BO at $91 was just not possible, because it takes at least 30 minutes, using Mike's Volume Alert calculations to get an alert.



    Therefore, we need to use $93 as a BO value for this strategy. Using a higher BO value automatically decreases the R/R ratio, because higher entry points means a price that is extended and that will revert back to hit the 5% stop loss.



    Hence, higher BO entry points require wider stops. For example, using a 7% stop level, we can see that the trade offers a better R/R.



    Fortunately for me, the price reverted back to its BO point of $91 on Friday, where I bought a position. Also, we can see on the EV Figure above that a natural stop is $88, which is tighter than the 5% stop. Finally, keeping the trade for 10 days allows also to improve the R/R and Strategy Efficiency.



    This tells us that:

    1. If we can get an early entry using Mike's Volume Alert system, then we will should do fine
    2. If we cannot get early enough and the stock is extended, then the stop must be wider OR we need to wait for a pull-back

    This new software will instantly give me the best stop to use OR the best pull-back price to wait for.

    Let's continue with PAA, A slow moving high-dividend Pipe-line company



    PAA is a leading stock. PAA attracted money on Friday, because of the general liquidity transfer between US Treasuries and Dividend offering equities. The question for us is simple: should be buy a break-out at $61, a BO on volume or a pull-back to the 5MA?



    We can see by comparing the three tables below that buying pull-backs is the preferred strategy for PAA and that the least interesting strategy is to buy BO on volume, probably because there are algos that calculate the value of AAP on the base of dividends expectations and hence quickly short-sell slightly extended value based stocks in order to balance existing long trades on cheaper dividend stocks.







    MSFT shows another type of results: Pullbacks have worked better than normal BO and BOV.










    Conclusions:

    I understand that any trader who reads this article will want to use the software.

    We can clearly see that it will be a great addition to Mike's Volume Trigger alert or to any stock picking service, because this software easily allows to find the best setting for each specific stock. This is what it takes to catch the best of an existing trend.

    My problem is that I would like to offer it, but do not want it to be distributed without authorisation or even reverse engineered, repackaged and sold. The software has been developed in Visual Basic, which is very light on the security side.

    I do not want to sell software either, because this requires maintenance, which is the reason why I stopped offering the EV plug-ins.

    As of now, the software is too new to be distributed. There is no user manual and I am sure that bugs are still left in the code.

    The best for now is that I continue to use it in the stock selection process and see that everything is ironed out. I will also examine with the tech guys how this software can be used while a subscription remains valid.