The bottom line, as mentioned in the last night report, the previous high is the key to watch. We did break above the previous high today but unfortunately it’s not decisive enough so guess we’ll have to wait for the next Monday. I have more detailed evil plan of course but I’ll leave it to the weekend’s report. Basically, I think another higher high is likely. It’s not important anyway because right now the most important thing is to enjoy your weekend!
A little bearish biased toward the next Monday because both VIX and SPX were up today.
Cobra Impulse System will try to add into the current long position the next Monday should SPY break above today’s high. So set SSO buy stop at $55.22, stop loss around $53.40. I’m not sure if it’s a good idea though especially if you still hold all the pyramided positions.
- A little attention to the chart below, now we finally have the roughly annual return so that you can compare with the old version of Cobra Impulse System, which was 17.1% for long and 14.12% for short, now is 25.30% for long, 16.46% for short and most importantly, I’m still working on the system, should be better later.
THE SHORT-TERM MODEL BELIEVES THE TREND IS UP, HOLDING NO POSITION OVER THE WEEKEND:
THE INTERMEDIATE-TERM MODEL BELIEVES THE TREND IS UP:
thx,ding
Thanks much for the update on the Cobra Impulse system. That is after all the main reason I come here. It works pretty good as an intermediate trend signal. Just curious though why you seem to sell many of the pyramid adds so quickly after only holding a few days before you get much profit. If intermediate term is looking for trends to last several weeks to maybe even several months then I just keep adding and holding especially under current market conditions with lower volatility. I have held and added all the way up per CIS adds from 12/21. At this time I am full and will not add any more unless we get a solid jump into a new wave up. Then I have one more load left.
But as you say, you just give the signals. It is up to us how to fully implement them.
Thanks again for all your work. It is appreciated.
Because I haven’t found a way to exit properly (I need program everything, don’t forget that, so even I have some good idea to exit, but it’s not easy to program that). Based on the current exit strategy, statistically, as soon as there’s a breakeven stop loss, out all,yields better results. Plus, you never know how long a trend would last, so in the long run, hold all the added position might not be a good idea, the statistics might prove correct.
Thanks for the CIS update, Cobra.
Two question: 1. Is the data on the system’s returns based on SSO (i.e., twice what one would expect if one invested in SPY)? 2. How far back is the system backtested to yield the data you presented today?
1. SPY.
2. 2,000 trading days. If apply 5,000 trading days, it’s even better.
Thanks, Cobra.
When you say that X% of long trades are profitable, how are trades calculated? Is a long call plus all its subsequent pyramid positions considered one trade, or is each pyramid position considered a separate trade? In other words, if, for example, you say that 85% of the long trades for the CIS are profitable, does that mean that, on average, each pyramid trade has a 85% chance of being successful?
For each trade.
Every time there’s an exit, I calculate the return and winning rate. Say:
1. $100, win $1, out $50, so the winner is 1, return 1%, you still have $50.
2. Later, $50 stopped out with $0.5 lose, so the loser is 1, return -1%.
The total winning rate is 1/2 = 50%. The total return is 1% + -1% = 0%.
The results after applying all the tech from old CIS:
The short setup still has to be optimized, not good enough.
Am I reading this correctly: the current CIS system now has backtested results of 31.60% annual profit for longs and 18.60% for shorts? Tell me my eyes are working properly and that’s what I’m seeing.
What you see is correct, assuming there’s no calculation bug in my program of course.
This is equal loss model, i.e. every time the loss is fixed, so when the volatility is very low, you might invest more than your capital.
The attached is the traditional way of invest, every trade invest $10,000, no matter the volatility. The results after 2,000 trading days is $11,211. So the actual return is:
11211/10000 / 2000 * 260 = 15%, instead of 31.60%.
So I guess, I should use the fixed capital model to be real because the fixed loss model (which is better by the way, I have a post discussing this) involves margins.
What do you think?
On 2nd thought, 11211/10000 is still not correct, because of pyramid.
Assume invest the 1st 10000, half out, still hold 5000. The pyramid would be another 10000, so the total capital is now 15000, there’re some cases the pyramid is 2, so holding 10000, adding new, it’d be 20000 instead of 10000 capital.
So I think the actual return is much much lower.
I do, except for one thing. I thought that the pyramiding process involved selling a position and then buying it again, so that, at most, one kept half the initial position and then one position that is bought, sold, bought, etc. In other words, say the initial investment is 100 shares of SPY. Then, let’s say that 50 are sold, so you have 50 left. The first pyramid would be 100 shares, leaving you wtih 150. After that, my undertanding is that you would sell the first pyramind (i.e., 100 shares), be left with 50, and then buy 100 shares again if there’s a second pyriamid.
However, in your response, you seem to say that pyramiding works by accumulating shares.
I’m guessing that the answer to my question is that shares are accumulated when a pyramid position isn’t stopped out and not accumulated when it is. Is this correct?
You understood correctly. In theory the max hold is 150 shares in the most time. But there’re some very special cases, the max hold is 200 shares, i.e. I allow pyramid twice.
Anyway, this pyramid stuff make things complicated. In the reality trading, it’s practical as long as your account allows 100% margin. Pyramid is designed to work with margin, in theory.
I mean in theory, I should only hold 50 shares to the end of the world. The 100 share added is just temporary, but still it’s your capital, so the max capital you invested in this case is 150 shares.
In allow pyramid twice case is:
buy 100 shares, out 50.
buy another 100 shares, out 50, now you have 100 shares left.
buy another 100 shares, if no chances to out, you actually hold 200 shares now.
In the reality trading, you might:
buy 100 shares, out 50.
buy another 100 shares, out 50.
buy another 100 shares, out 50.
buy another 100 shares, out 50.
… … until the end of the world. 🙂
OK, I figured out why 31.60%, I shouldn’t calculate this way. See image attached, so far the portfolio this year, only gains 0.65% (roughly of course), but the return I calculated based on my method is very high. Simply adding every return together doesn’t reflect the truth and it’s completely wrong. I have a single UCO trade that returns 18.41%, but it only sold 30 shares, but this 18.41% would make the total return very beautiful.
Test results for all the other assets. 10000 capital invested, treat the pyramid as using margin, so the total return can be calculated by dividing the final gains with 10000. I think this is the only way to get the rough idea about the return.
One thing I want you to know is, the system is just to test my idea, it proves working, that’s all. The reality trading should be better because you can decide when to out the pyramid.
Thanks for the reply. For long term stats yes it works good your way. Especially during the massive volatility of 2011. Then I would trade exactly like you do. But this year so far has much less volatility so doing more buy and hold techniques. I do not just rely on your signal alone. I also do lot of other analysis myself using a variety of techniques to confirm.
So I have two techniques. Yours for volatile periods and buy and hold longer in melt-up trends like now as verified by my MV buying selling volume analysis charts and others. Then just move stops up.
So I personally have to make the techniques decision. But this year massive profits so far using the buy and hold method. But I will sell some before you give a short or sell signal. But so far even the short term trend has not broken stride much at all.
Attached I have included my 1 day MV chart for SP500 that uses proprietary technology not found anywhere else except with this vendor. It so far clearly show we are starting a potential third wave in the melt-up process.
I also included this chart in case you have not seen this kind of technology available only through marketvolume.com
Have a nice weekend. Time to head out myself.
Thanks. I’ll take a look at your strategy. Hopefully I could figure out something new. 🙂
Again, thank you, Cobra, and I wish you a good weekend.
This question is simply a sign of my dimwittedness, but here goes. You say that, for example, the 2nd long is held until there’s a breakeven stop loss. If you sell to break even, then you would never make a profit. Clearly, I’m missing something.
2nd long held until a breakeven stop loss can be set, then out all. i.e. every time when the daily bar low is above the entry price, sell this position. It sound’s not logic ha, so I might change this later, depends on the back test of course.
I’m beginning to get the hang of this.
So, if one held the first long position until the system went short, would that produce significantly inferior results to using the pyramid methodology?
not significantly worse, around 22% yearly return for long (short has no pyramid), but still pyramid is better. I’ll try some algo to see if system can hold the pyramid for awhile instead of out all.
Thank you, Cobra. And, if it isn’t obvious already, I really appreciate the improvements on the CIS. And, I mean not just the improved backtested performance, but the buy stops. The latter make the system much more user friendly.
-1.02 Change
Next week something is going to happen.
http://stockcharts.com/h-sc/ui?s=$NYMO&p=D&yr=0&mn=9&dy=0&id=p53389953895&a=257094644&r=7215&cmd=print
weekly
http://stockcharts.com/h-sc/ui?s=$SPX&p=W&yr=2&mn=0&dy=15&id=p30626667982&a=253237160&r=2696&cmd=print
I only know the next Monday or Tuesday we’ll have big move.
Hi Cobra,
Would you please give us your thoughts on the direction of this big move? I’m in cash 100% over the weekend but I appreciate your advice so I would know what to prepare comes next week. for example: 80% up or 80% down?
Thanks
No idea. NYMO small move usually means a big move on the indices ahead but the direction is unknown.
Statistically, QQQ up 8 weeks in a row, the next week had 5 out 7 (71%) chances to close in red.
Thanks Cobra. I’m studying the charts before heading out to dinner and it looks toppy to me for a reason. Man, if they drop 1% or more on Monday I’m gonna miss out on the fun. Let’s hope the market goes up a bit more.
The 1376 level acts as a magnet for this market. However, we might have topped on Friday at 1368..