Black Boxes, Mental Models, and Foolish Fours (Fool on the Hill) March 2, 2000

An Investment Opinion

Black Boxes, Mental Models, and Foolish Fours

By Bill Barker (TMF Max)
March 2, 2000

Clicking around cyberspace the other day, I came across a press release put out by, announcing that it had made Forbes' "Best of the Web" list. According to the release,'s selection was "due to its Research Wizards, which include a group of computerized financial minds, based around world-famous investors such as Peter Lynch, William O'Neil, the Motley Fool, and Ben Graham."

Intrigued at what the computerized mind of The Motley Fool might look like (would it be wearing one of the funny hats?), and figuring that it was kind of a compliment for my employer to be included with such legitimate investment luminaries, I decided to check out the website and kick its tires a little bit. Under a portion of the site called "Guru Analysis" I found the area that would apparently tell you how certain stocks fared under the computerized financial mind that was meant to mimic The Motley Fool's thinking. You can enter a stock ticker and the Guru Analysis will tell you whether the company is of "Strong Interest," "Some Interest," or "No Interest" according to the computerized methodology for the various "gurus."

I tried various real holdings of the Rule Breaker Portfolio and the Rule Maker Portfolio. I entered Microsoft (Nasdaq: MSFT), (Nasdaq: AMZN), Celera Genomics Group (NYSE: CRA), Cisco (Nasdaq: CSCO), Coca-Cola (NYSE: KO), JDS Uniphase (Nasdaq: JDSU), America Online (NYSE: AOL), Yahoo! (Nasdaq: YHOO), Intel (Nasdaq: INTC), and Starbucks (Nasdaq: SBUX).

For every one of them -- every one -- the computerized mind said The Motley Fool "has no interest" in those stocks. In fact, for every one of the companies, the model concluded that The Motley Fool might be interested in shorting the stock.

So how did that come about?

Well, one of the problems here is that black box models won't really tell you very well what a real person will do. I'm not blaming here; the methodology that the "computerized mind" uses rather faithfully employs the stock selection criteria set forth in The Motley Fool Investment Guide. However the specific criteria that were mentioned in that book were pretty much dedicated to picking small-cap companies, and the book was written more than four years ago. Things change. Thinking evolves, sometimes dramatically, as it should.

The market has changed too, or at least it has in some people's opinions. Sure, I can find plenty of people that are writing that the correct method for how to pick good companies hasn't changed at all. You just find the ones with P/Es below 10, and if there aren't any of those, everything is overvalued and there must be a BubbleMassHysteriaLemmingInvasionTulipMania going on. Or something like that.

The better thinking that I read out there isn't making that argument at all. Let me follow up on Whitney Tilson's reference in Monday's Boring Port to the fine work at CAP@Columbia, and specifically to the article "The Triumph of Bits." While Whitney sources it for its treatment of "Competitive Advantage Period," the subtitle of the article is "Mental Models for Successful Investment." The introduction to the article is key:

"Investing is a context-dependent activity. Change is inevitable. Rules of thumb that accurately represented the past have lost their validity. Investors more concerned with consistency of approach than evolution of approach are falling out of synch with the markets.... We believe the mental models required for the 21st century are very different from those that dominated even one generation ago. Investors today must adapt, as they have in the past. History has shown that those quickest to adapt reap disproportionately higher rewards. Change is the only constant."

How true. Whether you like 'em, hate 'em, or have never heard of 'em before, the Rule Breaker and Rule Maker ports have been very well served by their evolutionary mental model behavior. The methodology for picking stocks in those ports has very little to do with the criteria that are set forth in The Motley Fool Investment Guide and still presented as "the Motley Fool thinking" by the black box computer model.

Now the criteria in the Investment Guide might or might not still be very useful for an investor, and some of the principles are still very much involved in the stock selection criteria of each port. But the evolution away from many of the earlier criteria, the PEG or "Fool Ratio" in particular, has been dramatic. It is the high Fool Ratios of virtually every stock held in those two ports that is what has the black box of The Motley Fool mind recommending that those stocks might be short candidates.

Compare the successful changing mental models to the local portfolio that has not really evolved, itself a black box methodology. The Foolish Four portfolio is currently suffering some underperformance that has not gone unnoticed in the community, nor within the portfolio write-ups themselves. Now there are a lot of people from the community posting on the message boards good, solid questions about the Foolish Four selection methodology and historical performance data, and there are a lot of others asking whether the Fool Four "still works."

Count me as one of those who thinks that it is probable that the Fool Four does not "still work," (whether it has ever worked is an independent question and well presented by some tenacious gadflies). It seems to me it is not only possible but probable that no black box methodology is going to be a successful market-beating strategy over decades. As to this point, I think the pursuit of consistency is the hobgoblin of small minds.

The debate about the Fool Four will rage on for the time being, and my opinion is that those in the community raising questions about the validity of the theory will ultimately triumph in their pursuit of the issue. To that I have two thoughts -- one is that there could be no better demonstration of "Learning Together" than for the community posters to prevail, and the second is that it will be additional evidence that "the mental models required for the 21st century are very different from those that dominated even one generation ago."

Or even a couple of years ago.