The BMW Method
Confessions of a Lurker

Related Links
Discussion Boards

By tlpriest
September 27, 2006

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

I've lurked here for almost six months now and this is my first post to the board. Because the traffic here is so high, I generally only read the highest rated posts and threads, so I apologize in advance if anything I cover in this post has already been discussed. The BMW Method makes a lot of sense to me and I've been "kicking the tires" ever since I stumbled across it from a mention on another board. Up front, I'd like to thank BuildMWell for sharing the method, Mike Klein for his web site and related materials, and Jeff (real name?) for his stockreportv9_test spreadsheet. I believe these materials, along with the discussions on this board, are beginning to make me a better investor.

I have a portfolio of seven stocks (Peter Lynch). In the last year, six of them have experienced highs or lows that, in retrospect, I wish I would have done, or not done, something about at that time. The BMW Method correctly signaled all those times. Here's why I figure that to be true.

First, I'll let you know that while I was very strong mathematically in college, I failed statistics the first time around, mostly because I took it in a six-week summer session and skipped half the classes. Obviously, I didn't take it seriously. I didn't do much better the second time because it was non-core and I was ready to get out of school for my job. I regret that blas� attitude now! With that in mind, I dissected Jeff's spreadsheet to figure out how he calculated the lines of standard deviation on the plots in that spreadsheet (I remember standard deviation!) by lifting his work into a spreadsheet of my own and walking through it line-by-line several times. The MS Excel help system and the web helped, too.

I generally pull daily price history data from yahoo (for other financial spreadsheets I've created), so after massaging his code, I got acceptably close answers to his monthly data from MSN. Since all I wanted was something that does a fast BMW analysis, I didn't add anything else from his spreadsheet (although there's an amazing amount of good stuff in it). That sure makes updates a lot faster!

I did add the ability to select start and stop dates for the analysis. I think start dates are very important for BMW analysis, especially in mid- and small-cap stocks, because some companies may go through events that fundamentally change their business. Price trend analysis should start at that point, and not further back in history, when they were a "different" company. I own or have owned several of these: MVL, which came out of bankruptcy in '01 (off the top of my head) and started focusing on licensing its stable of properties to Hollywood, FARO in '01 (again, off the top of my head), when it changed its focus to metrology, and QSII in '99 when it acquired the company that gave it Nextgen. I've looked at a lot of BMW graphs now, and I often find it useful to look at the time period where the price hovered under SD-2 for a clue as to whether it was a "life-changing" event � did they recover to their previous CAGR or establish a new one? Being able to change the start time of the plots also helps in determining if the current price is a short-term or long-term trend, and to compare CAGR's in those time periods.

Having a stop date is useful for back-testing. If you look at a chart generated today and try to determine if it was accurate a year ago, you have an unfair advantage. The lines of standard deviation and CAGR are affected by the last 12 months of price data, so they can paint a different picture than what you saw a year ago if you'd been viewing that chart for the first time. So, if you stop the chart a year ago for a "snapshot back in time," you have the benefit of knowing what's happened in the following twelve months while only seeing what you would have seen back when it was happening. I've learned a lot by looking at charts of companies whose events I'm familiar with the snapshots back in time. I made it so that I could step forward month by month, which illustrates how the CAGR changes over time and to easily compare what I knew then with what I know now. That knowledge has removed some of the psychology that's plagued me.

Finally, I got rid of the calculations for split+dividend adjusted prices � while I recognize dividends are an important part of the total return picture, I found that it provided me with too much to look at and it was too difficult for me to track. KISS.

So, where has the BMW Method been right for me in the last year based on this work? A few examples:

QSII makes up a disproportionately large part of my portfolio (woo hoo!), and I should have bought protective puts last year when the price hit $88 ($44 split-adjusted). It was SD+2 at the time. Just the month before I'd finished the options authorization paperwork on all my brokerage accounts because I was concerned about QSII's valuation (I wanted to hold it, but I wanted some insurance, too), but didn't know when to pull the trigger. I missed. BMW knew, though. Plot start: 4/1/99 (That drop from $90 to $60 hurt like @#$%^&*!)

MVL's management leaves a lot to be desired in my opinion. I sold it a few months ago in disgust with the changes (insider sales, new movie studio, Avi's shenanigans). Of course, I sold it under SD-2 because I didn't check to see where it was on the BMW roadmap. That was a short-term mistake that BMW could have made me think more about at the time. Plot start: 1/1/01

COP was an issue that I'd actively thought about selling when it was in the upper 60's recently. I got really busy (adopting a baby girl from China) and put the sale on the back burner. A quick gander at the BMW chart could have triggered me to rate that sell a higher priority on my todo list before I left on my trip, since it was at SD+2 (starting the plot from 1/1/01 � we can argue if that's a valid thing to do later).

There are other examples, but you get the idea. Anecdotally, BMW seems to work really well on my portfolio. It's reassuring to have another measure of where you might be on the value road.

With the uncertainty in the economy, I'm feeling the need to look for value a little more and give growth a break for awhile. I like TMF newsletters because there are a number of eyes on their picks at any given time, they fully explain the investment thesis, you can get multiple viewpoints, and if your question is good enough, it'll be addressed by the staff in a newsletter or taken to the company itself. This is more analysis than I can do on my own, although I do make my own picks outside of TMF about 50% of the time. But, I've been curious to see where the TMF picks rate on the BMW value charts, so I added a macro to my (mostly stolen, thanks again!) spreadsheet code to loop through the list of 194 SA & HG picks and record their present-day RF and CAGR values.

I've come up with three picks (all from SA) that are priced well right now according to BMW but don't have terribly obvious "flaws." I don't know what TMF's policy is about revealing newsletter picks on non-newsletter boards, but I think you can easily glean that one in particular, my favorite, is an SA pick without subscribing to the service.

So, I'll throw out BBBY's ticker symbol here. It was obviously better-priced two months ago, but it still sits below SD-2 for no discernable good reason that I could find that's specific to this stock other than the housing market. What I like even more about the BMW analysis on BBBY is that its 14-year CAGR is 27%, but its 6.5 year CAGR is only 11%. Using the more conservative (CAGR-wise) short-term chart sets the current price just 10% over the SD-2 line. Fundamentally, it looks great � good enough to add to my portfolio last week.

I've also thought about looking at the AAII screens � Zweig, CANSLIM, etc. � (ok, arguably not the value I mentioned above) and taking out the price appreciation elements of those screens (near 52-week high, increasing volume, etc.) and just using the fundamental parts of those screens combined with BMW to see if there's anything interesting. Has anyone done this or thought about doing this?


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.