As it turns out, adding the CCI system into the mix and creating the JT Composite did not work out so well this year. Because the CCI underperformed vs. the original system (now known as the JTS) the JT Composite also underperformed.
Stats will be broken down into 3 mini-sections below. The first will simply look at 2012 returns for the CCI, JTS, and JT Composite. I will also include profit curves for each. In the next section, we'll look at the the 2012 returns for the JT Composite 100% strategy. Finally, I'll show all-time JT Composite stats dating back to Jan 2009 and how they compare to the all-time 100% strategy.
2012 CCI Stats
28 winners, 16 losers - 64% win rate
Avg. Loss: 6.22%
Avg. Win: 4.89%
Profit Factor 1.38
2012 JTS Stats
36 winners, 13 losers - 73% win rate
Avg. Loss: 3.88%
Avg. Win: 3.58%
Profit Factor 2.56
2012 JT Composite Stats
44 winners, 30 losers - 60% win rate
Avg. Loss: 2.70%
Avg. Win: 3.14%
Profit Factor 1.70
For the sake of comparison, here are the stats using a 100% strategy, only going long or short when both the JTS and CCI are in agreement. (Notice, still not as good as JTS. Interesting...)
2012 100% Strategy JT Composite Stats
27 winners, 13 losers - 68% win rate
Avg. Loss: 3.44%
Avg. Win: 3.98%
Profit Factor 2.40
Finally, here are the all-time JT Composite stats spanning from early 2009 to 2012. The JT Comp cannot go back any further b/c the JTS uses TNA in the calculations, which came into existence 4 years ago. The stats suggest the JT Composite underperformed this year vs. the previous 3 years.
All-time JT Composite Stats
201 winners, 99 losers - 67% win rate
Avg. Loss: 4.05%
Avg. Win: 4.57%
Profit Factor 2.29
Last but not least, the all-time 100% strategy JT Composite system.
All-time 100% Strategy JT Composite Stats
112 winners, 36 losers - 76% win rate
Avg. Loss: 5.24%
Avg. Win: 5.96%
Profit Factor 3.54
My conclusion (yours may vary) is that the JT Composite is a solid system. However, it can be improved upon by following it only when both the CCI and JTS are in agreement (100% strategy). This requires that you have the patience to pull this off, as there could be lengthy periods on the sidelines. (I'm training myself for this by taking the month of January off. :) The other conclusion is that the JTS remains king. I'm not giving up on the CCI or the JT Composite b/c I know the CCI can perform remarkably. I've seen it firsthand. I will put the CCI on a tight leash, though. If it doesn't perform better this year, I will consider shelving it or minimizing it in JT Comp calculations. My grand takeaway after looking at this is that I will only be following the 100% strategy this year. Another option is to simply follow the JTS, but I prefer having the CCI incorporated.
Thanks for reading. Good luck to us all in 2013!!! I'll be back in the game Jan 28th. Take care.
J
PS- I'll add a couple goodies this weekend including a chart of my 2012 trading and a spreadsheet with all system trades for 2012.
Thank you very much, J. This is very informative and gives much food for thought.
ReplyDeleteI have two questions:
1. The all-time JT Comp equity curve appears to be relatively flat for about two years and then it shoots up. Do you have any idea why that might be?
2. Did you have a chance to see how the 100% JT Comp would have done using TNA?
Once again, many thanks for your hard work and generosity.
Your first question simply boils down to a scaling issue. The starting value is a mere $10,000 so appearances in this case are deceiving. During the time that the curve appears flat, it's actually growing from 10,000 to 100,000, which is obviously nothing to be too concerned about.
DeleteI just noticed that your stats are for the 100% JT Comp do use TNA. Sorry for my confusion.
ReplyDeleteIf you have the chance, it would be interesting to see the difference between the CAGR for the SPY and TNA systems during the longer test period.
Thanks.
So if I understand correctly the graphs represent compounded gains? But you do not use compounding in your regular system returns on the left of the page, correct?
ReplyDeleteSomething useful might be to compare 100% JTS with 100% JT Comp all-time. I wonder if JTS is more profitable but also more volatile for the entire period.
ReplyDelete