Review of the VIX & More Stock of the Week


This is a review of the VIX & More Stock of the Week included as part of V&M’s subscriber newsletter. As the name implies, Bill Luby (the brain behind V&M) picks a single stock each week and holds it until the following week, and has issued 107 picks since the newsletter’s inception.

The SOTW isn’t independently-audited in the strict sense, but because of the nature of the newsletter (issued just once-a-week before the trade) and a large readership keeping Bill honest, I feel good about using his list of trades in this analysis.

[logarithmically-scaled, frictionless, weekly-interval]

Bill’s newsletter is issued every weekend. The graph above assumes a subscriber purchased the SOTW at the first open and sold at the last close of the following week. The SOTW is in red, and for comparison, the SPY is in grey.

A couple of important notes about these results:

First, these results are frictionless (i.e. do not account for transaction costs or slippage) because trading frictions are very specific to the individual trader. The average/median return per trade was 1.38%/0.89%, so there’s quite a bit of meat to work with.

Second, these results (while very good) are south of what Bill uses for tracking purposes in the newsletter. Newsletter results assume a trader purchased the SOTW at the last close of the previous week (before the newsletter is issued). Bill has always been very clear about this and hopefully the chart above helps subscribers better understand how a delayed entry affects the program.

Numbers for the number lovers…

Thoughts on the VIX & More Stock of the Week

Bill has shown a real knack for picking weekly winners on a fairly consistent basis.

Periods of flat returns have generally coincided with bearish periods in the market (which is to be expected with a long-only equity program that’s always in the market).

I’ve assumed a trader simply bought at the open and sold at the close, so there are all sorts of opportunities here for intraday traders (which I am not) to improve entries/exits.

One last note: it should go without saying (but I’ll say it anyways), because of how concentrated the SOTW portfolio is (just one stock), it should only constitute a very small portion of a trader’s portfolio.

The Play Money Project

In 2009 I decided to commit $100k to 10 test accounts trading all sorts of programs that (a) looked promising, (b) could be traded mechanically by a small investor, and (c) were radically different than my own. I’m calling it my Play Money Project.

This is the first in what I hope will be a series of posts talking about the successes of the Play Money Project, but note that I still haven’t found all of my 10, so if you have any suggestions please let me know.

[click for a summary of all recent posts about the VIX & More SOTW]

Happy Trading,

P.S. I am NOT being compensated in any way for this post – just trying to shine the light on the good guys.

. . . . .

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.

13 Responses to “Review of the VIX & More Stock of the Week”

  1. 1 Mike

    Michael I’d love to see your analysis of some of scott’s other profunds offerings. There are several with promising short term results. In particular it would be nice to see a beta analysis with yk as a nice way to round out a portfolio with low correlation but solid returns.

    • 2 MarketSci

      RE to Mike: I would love to but I try to minimize the for-profit stuff on the blog. I’m afraid it will look like cherry-picking if I highlight good programs beyond my own (and I am ethically opposed to cherry-picking returns). michael

  2. 3 Paul J

    [Excuse me, but this is a correction of a previous comment.]

    Thank you for reviewing the Stock of the Week portfolio.

    To reduce risk and volatility of a single stock portfolio, one may consider hedging it with a short position in S&P 500.

    I think that that would be worthy of further analysis.

    Thank you very much, and keep up the great work.

    • 4 MarketSci

      RE to PaulJ: you stole my thunder…that’s exactly what my next post was going to be about…michael

  3. 5 Roger


    How about looking at some the CSSAnalytics/DVIndicators stuff. I’m pretty impressed with mean reversion indicators like his DVO which managed to outperform the market in 2009 (at least backtested). I think few mean reversion system/indicators outperformed in 2009 (or at least the latter half) – it was all about trend following.


    • 6 MarketSci

      Hello Roger – good thought, but the Play Money Project is intended for complete programs being traded in real-time, not indicators that would be subject to interpretation. In other words, it has to be something that someone is offering that could be followed mechanically. michael

  4. 7 Roger

    One more idea – how about some of the indicators from BZB Trader. I have coded some of these and have seen they perform well in the recent market but not so well if back tested for longer periods. So the question is – am i doing something wrong (entirely possible) or are they optimized for the current market (which is an interesting consideration – is an adaptive system approach better (go with what is working recenly) than a long backtest? Or how about a self adapting system that backtests well e.g. I believe this is how YK works – although YK has not been doing well lately as I’m sure you are painfully aware.)

  5. 8 Paul J

    Hope I’m not stealing your thunder again here (lol), but…

    How about building a portfolio of say, 10 stocks using SOTW selections. It could be 10, 5, 11, 2, whatever. Actually, it would be good to see several numbers of stocks tested.
    For example, every stock of the week stays in the portfolio for 10 weeks, then the oldest one is replaced by the latest pick.
    Sure, this really changes the portfolio, but since Bill’s picks are pretty good, it would be interesting to see how his picks do over a longer period.
    This provides diversification in the portfolio. It would be interesting to see how that affects return and volatility.

    Thank you…

    • 9 MarketSci

      RE to PaulJ: now you’re stealing Bill Luby’s thunder =) He’s mentioned that he is considering doing just that – running a 10 stock portfolio w/ 1 stock being rotated out each week. I might ask Bill if he minds if I run some numbers on that b/c I already have the data together. More to follow. michael

      • First, thanks Michael for putting the SOTW through the paces here. The interest in your effort and questions raised by all the discussion will certainly help me come up with a better product — not to mention some better statistics to explain what has and has not happened since I started this effort.

        Paul’s idea is exactly where I was headed here. I just haven’t been able to make the time to do the historical calculations, etc. and set a rolling 10-stock SOTW portfolio in motion. By all means, if someone else wants to do something similar in terms of historical statistics, I would certainly encourage that. It might also help to accelerate my timetable. As a one man show with an overflowing R&D inbox, I am clearly in need of some better cloning technology.

        Cheers and thanks to all who have weighed in on the SOTW,


  6. 11 ashpak001

    What about running some Portfolio123 strategies with backtested AR of 50-80% in real time.

    • 12 MarketSci

      RE to ashpak001: will check it out – thanks for that. What does “backtested in realtime” mean? Do you mean hypothetical execution walked forward? michael

      • 13 ashpak001

        Sorry – typo – I meant running in real time backtested strategies :)

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s