Actually results would have been slightly better than the results of Harris's patterns. You've won the argument, but you won't be able to convince Bill. Harris choose his long pattern "out of sample" period with the benefit of hindsight. The patterns did slightly worse than would have been expected due to chance alone. Bill has little formal statistical training, after dropping out of Tulane in his freshman year, he ended up with a Bachelors from LSU-Shreveport, a bottom tier school. He continues to shill for Harris and his APS program, on this and other forums, despite ample eveidence that the program is worthless. AFAIK, neither Bill nor Harris have published in any recognised peer-reviewed journals.
Googling for Bill + markets + trading + Tulane + "LSU-Shreveport" I found the following. Is this the same Bill you're talking about? This one seems to have been an actuary once, which requires some amount of statistical proficiency. And this one seems to be a buy-and-hold investor rather than an intraday trader. Bill graduated from Caddo Magnet High School (a high school for nerds) back in 1985 and proceeded to learn the hard way when he drank his way out of a scholarship to Tulane later that year. After a few years of sweating for a living, he decided to go back to school, and graduated from LSU-Shreveport in 1995 with a Bachelors in Mathematics - all the while working the overnight shift stocking shelves in a grocery store. Post-college, Bill has been in the P&C insurance industry as an actuary, product manager, and pricing manager. Bill and his wife Millie are amateur investors with a variety of holdings, but they prefer to buy and hold value investments.
This is what Bill wrote about APS in another thread and actually he tried to discourage someone from buying the program. So Bill agrees that most systems these programs find do not have an edge. I think his argument was about something else.
We will see who is lying through official routes if you continue making false accusations. I did not have any time to respond because I actually trade, unlike you. I am posting the results of 4 random systems I tested using Metastock for the same period as Michael Harris did the out_of_sample testing, 05/07/2002 to 08/22/2008 in this paper. When I saw that the results did not agree with your alleged findings which you defended using insulting manners and accusations, I also lowered the R:R to even get bigger samples that would increase the applicability of your "large collections" argument. No matter how many random test I run, I was not able to get any close to the profit factor in the article. The highest I got was 1.2 but this is not the issue. Probably one could find a random system with a profit factor greater than that of the system in the article. But that is little related to system trading. Even a dummy can understand that, it appears you do not. Unlike you, I am posting detailed results here of four such random systems. Random system 1: PF < 1 win rate WR = 49% Random system 2: PF = 1.03 WR = 52% Random system 3: PF = 1.2 WR = 55% Random System 4: PF = 1.04 WR = 52% I attach the backtesting screenshots from Metastock. Beware that your argument that any random system would have done better than the system APS found has been demolished. Keep in mind that I did not take in account commissions and that would impact adversely large samples. Find a better argument. If you come back with insults, you will face the consequences and the same holds with your other collaborator, Emilio Lizardo.
First of all, I'm not your babysitter and won't reply to your garbage infinitely. Now, moving on. You just lied again. Or maybe you are too dense to realize the difference between any and average. At this point, I'm not sure what's more likely. About your pseudo-analysis. Posting individual results of a few cherry-picked random runs is pointless because individual random trade selections can't be repeated and verified. I always made it clear that the overall performance of a large number of random trade selections averages to 60.5 +- 0.5 winning percentage. That's a verifiable statement. The obvious and huge mistake you made â and you are either too stupid to realize it, or you are deliberately deceiving again â is that your trades don't overlap. You wrongly wait until your stop/target gets hit before you enter again. So you didn't weight all days equally like I did, your entries are biased and your results are screwed. Keep in mind that Harris' 8 patterns combined can have 8 open positions at once, totally independent from each other.
You are constantly accusing me of lying. Are you willing to prove this in a court of law? People trade individual systems. You just averaged so many samples so that your result will be the buy_and_hold return at a huge commission cost. Before you accuse people you must be very careful. It may cost you a lot. You have now invented a new straw man that the trades do not overlap. I have shown to you 4 random systems that do not exceed Harris returns and you are accusing me that I manufactured the results. Are you willing to go to court in front of an expert witness and defend your statements and accusations? You have accused me publicly of lying and you are going to face the consequences of that.