Message Font: Serif | Sans-Serif
No. of Recommendations: 17
This is my first post on this board. Not sure if you welcome outsiders or not, but I thought I'd test the waters a bit. I'm exploring some methods for conservative strategies. I have not had time to delve into this very deeply, but I'm quite familiar with Ken's work and the overall concept. In order to test this I used the picks that Lee has been graciously posting here for some time -- from late-2001 I believe. I basically pulled in the month end picks for 2004 to present on his two main posts of Midcap picks. The problem is, I do not really understand the difference between the two.
MidCap 400 BI and All MidCap BI Picks

There is a vast difference in the returns between the two, at least for 2004 to present. I understand that the All post is all MidCap 400 stocks under either 10 Yr downside or 5 Yr downside targets.

The other one seems to be just those that were updated last week, but is that it? If that is the case, I don't really need these picks for backtesting. Correct?

I am especially interested in the Piotroski values added recently. For this reason, I ran some tests where I did the following. I took the top 25 picks for each of the posts and called them per the names on the left. I then took the full list from the post that had the Piotroski value, and resorted using that column as the basis in descending order. The thought would be that these stocks would do best. I then traded once a month -- the first day of the month each time, taking the picks from the last post. I ran tests holding anywhere from one to twenty-five. The averages follow in this table:
    Aug 1 to Feb 18, 2005     # to Hold    ROI    CAGR   GSD(D)  Sharpe  Drawdown  Win 1  Win 2
S&P Midcap 400 Piotroski Avg 12.69% 24.57% 18.33 1.27 -10.97% 75% 65%
S&P Midcap 400 Piotroski All Avg 18.91% 37.36% 19.02 1.74 -8.32% 67% 65%
S&P Midcap 400 All Avg 23.73% 47.66% 17.78 2.38 -6.47% 77% 65%
S&P Midcap 400 BI Plays Avg 18.82% 37.19% 18.47 1.76 -7.45% 74% 65%

Jan 1, 2000 to Feb 18, 2005
S&P Midcap 400 All Avg 50.75% 43.71% 17.59 2.20 -8.28% 72% 63%
S&P Midcap 400 BI Plays Avg 21.22% 18.52% 18.36 0.96 -11.84% 65% 57%

Win 1 = Percent of combined rebalance results that are positive
Win 2 = Percent of individual stock picks that are positive

My questions are these:

1. Do these values look about right? Using the All, which is just the basic sort produces the best returns by far.
2. Why is it that in this short backtest the Piotroski values are not helping more?
3. Is there a source of picks farther back in time? It would be good to see how this did in 2000 to 2001, and earlier, if at all possible.
4. What is the length of the longest backtest on the Mid-caps?
5. What is considered the optimal rebalance date? In other words, how important is the freshness of the data used in the opinion of those who have used the method for a few years?

I tested the All picks and came up with the following results (the # held was just 5 to 10 on this test, not 1 to 25 as in the previous test):
  Rebalance   # to Hold    ROI    CAGR   GSD(D)  Sharpe  Drawdown  Win 1  Win 2
Annually Avg 44.34% 38.34% 17.68 1.96 -8.94% 83% 78%
Bi-Annually Avg 59.64% 51.23% 16.86 2.59 -8.30% 100% 91%
Tri-Annually Avg 55.55% 47.80% 16.01 2.56 -8.36% 100% 83%
Quarterly Avg 44.38% 38.37% 16.95 2.02 -10.30% 73% 71%
6x Annually Avg 45.48% 39.30% 17.08 2.05 -7.99% 90% 69%

Would you say this is representative of the longer backtest?

I can post all the results for each test with the various holds -- 1 to 25 -- if there is interest. I find it quite amazing the stability of the returns, especially for the All picks. Of course, what I've done here is an extremely short backtest and not worth much of anything, but I share it just to ask whether it is representative or not.
Print the post  


When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.