Jump to content



Photo

AmericanBulls.com 2 year results for GDX with expectancy


  • Please log in to reply
4 replies to this topic

#1 cometfish

cometfish

    Member

  • Traders-Talk User
  • 4 posts

Posted 29 October 2009 - 11:24 PM

I did some backtesting using AmericanBulls.com's daily signal for GDX from http://www.americanb...p...=NYSE&Typ=S.

Entry and exit are next day open. Using $100 original investment on 10/05/07, see attached comparison between stated result ($631) vs. actual long-only result ($144.26) vs. actual long-and-short result ($143.52). For long-only, the expectancy is 0.32, and for long-and-short, the expectancy is only 0.11.

Expectancy = (Net Profit / # of Trades) / Average Loss --- based on Van Tharp's formula

Make your own judgement ;)
-cometfish

Attached Thumbnails

  • AmBullsGDX102909.jpg


#2 milbank

milbank

    Member

  • TT Patron+
  • 4,714 posts

Posted 30 October 2009 - 08:51 AM

You beat me to something I was going to try but, never got around to. I was going to use some small cap like NVEC and see if my results and theirs were similar. Never got around to it. What a big difference you got. Thanks for posting the results and thanks for sharing your work using their advice.

"The power of accurate observation is commonly called cynicism by those who have not got it."
--George Bernard Shaw


"None are so hopelessly enslaved as those who falsely believe they are free."
--Johann Wolfgang von Goethe


#3 diogenes227

diogenes227

    Member

  • TT Patron+
  • 5,120 posts

Posted 30 October 2009 - 10:49 AM

Thanks for doing this. I've always been skeptical of their results partly because they seem too good to be true and primarily because their actual stated entry points are crazy -- buy if it's above the open? Well, you're not going to know if that's true until the close. Did you happen to test this out with the close as the entry point on days when their "buy if" above-the-open criteria is actually in place at the close? Thanks again for the work and best to you. :)

"If you've heard this story before, don't stop me because I'd like to hear it again," Groucho Marx (on market history?).

“I've learned in options trading simple is best and the obvious is often the most elusive to recognize.”

 

"The god of trading rewards persistence, experience and discipline, and absolutely nothing else."


#4 cometfish

cometfish

    Member

  • Traders-Talk User
  • 4 posts

Posted 31 October 2009 - 10:32 PM

Following diogenes' advice, I put aside my laziness and did the backtesting again using the signal day's close price. Here're the 2-year results: Long only: Total profit: 90%. Expectancy: 0.64 Long & Short (50% margin): Total profit: 92%. Expectancy: 0.22 So the performance is much better than using the next day's open price, although still way below their stated numbers. 0.64 expectancy is a good starting point. Combining it with some other longer term signal, and I think we may find something interesting.

#5 diogenes227

diogenes227

    Member

  • TT Patron+
  • 5,120 posts

Posted 01 November 2009 - 06:56 PM

Following diogenes' advice, I put aside my laziness and did the backtesting again using the signal day's close price. Here're the 2-year results:

Long only:
Total profit: 90%. Expectancy: 0.64

Long & Short (50% margin):
Total profit: 92%. Expectancy: 0.22

So the performance is much better than using the next day's open price, although still way below their stated numbers.

0.64 expectancy is a good starting point. Combining it with some other longer term signal, and I think we may find something interesting.


Thanks for running those numbers. I don't know why they want to fudge their results -- 90 percent on the GDX is not bad.

As skeptical as I've been on their results because of their cockeyed use of the open entry criteria on the day of the buy and sell signals (which is virtually impossible to trade), I think there is something to using candlesticks the way they try to use them (but using the close as the entry point). Since you ran these numbers, I decided to get over a little of my laziness (but not all of it :) ) and quickly crunch a few numbers.

I took a look a LVS because I like the way that stock moves, and AAPL because everyone around here seems to hate it.

Using American Bulls buying and selling criteria on LVS would have yielded 13.8 points of profit on four long trades since 7/14. Great, but impossible to match in reality. Taking the same signals on the end of the day, which would have been easy to do, yielded 9.94 points on the same four trades. The first trade would have been at 7.14 so those 9.94 points represented 138 percent profit. One hundred and thirty-eight percent!

Using the American Bulls criteria on AAPL gained 58.59 points according to them on five trades since 6/25. Not bad. Using the end of the day to buy and sell AAPL instead yielded 59.19 points. What do you know, about the same.

Those two stocks were chosen at random and results like these may not hold up across the entire market but it looks like these guys at American Bulls are so busy talking pie in the sky they might not know they actually may have a real pie in the oven.

:lol:

"If you've heard this story before, don't stop me because I'd like to hear it again," Groucho Marx (on market history?).

“I've learned in options trading simple is best and the obvious is often the most elusive to recognize.”

 

"The god of trading rewards persistence, experience and discipline, and absolutely nothing else."