It was, like, totally random

May 9, 2009 at 4:06 pm | Posted in World of Ideas | Leave a comment
Tags: , , , ,

Wednesday night, Tom and I went to see a talk by Leonard Mlodinow, from Caltech, whose recent book, The Drunkard’s Walk, is in bookstores and available for purchase.

I haven’t read this one, but based on his talk, I suspect it would cover similar ideas to the book A Random Walk Down Wall Street, by Burton Malkiel, which I read a couple of years ago.

The “drunkard’s walk” refers to the notion that, as with someone really drunk who is wandering aimlessly about, things that are truly random cannot be predicted. Though there are some stretches where there appears to be a pattern to the progress, even these seeming “trends” are, in fact, just the results of cumulative happenstance. As with the monkeys and typewriters notion, sometimes even random processes will line up into apparently coherent and consistent strings. Sometimes even the random hitting of keys will result in recognizable words.

The Malkiel book applied this notion to the stock market for the most part, pointing out that while there was little evidence that attempts at pattern recognition in stock trends were actually borne out as being effective, there is plenty to support the notion that in fact the stock market moves randomly, like the drunkard out on an aimless stroll: he may know his general direction, but it’s difficult to predict, from one step to the next, where he will go. In other words, he felt that things like day trading were little more than guess work. And, while you can see certain overall trends, like the current widespread, cross-sector downturn, it’s impossible to predict when it will turn around, by how much and so on. His conclusion was that mutual funds are to be avoided, and that an investor’s best bet is to just go with an index fund, using something like the S&P 500.

Derivatives might well change things up a bit. A money manager who has figured out a clever way to use those to mitigate risk might be able to come out ahead more consistently–but I only say that because I haven’t read enough about the range and variety of derivatives, how such instruments would work, and whether there’s actually evidence that they do.

All the same, as regards the markets in general, I found the arguments pretty compelling. Mlodinow branched off into a couple of additional facets of the randomness concept, but otherwise covered similar ground. I’d certainly recommend the Malkiel as readable, interesting food for thought. If the Mlodinow is similar (and possibly wider-ranging, rather than focussing on how randomness relates to the stock market), then it might be worth checking out as well. Whether you end up agreeing or not, it’s interesting stuff. If you happen to have read either of these books, but remain a staunch proponent of technical or fundamental analysis, I’d love to hear your arguments against the evidence produced and in favour of your approach!

For the record, I’d actually still do fundamental analysis and likely even do a bit of “tech” analysis too–in that I’d look at the general mood of investors, to try to gauge whether it’s a good time to buy and whether the stock is close to bottoming out, or reaching its peak, before investing. It’s one thing to agree that the day-to-day movements are generally random, and another to throw out the whole idea of a downward or upward trend in markets. As Malkiel points out, if you have a long horizon (i.e. 20-30 years before you retire), then just investing in an index should do you fine. It goes up over the long term, generally performing slightly better than inflation, and so your money will grow.

But keeping aware of trends, wading in when the market bottoms out (or at least is considerably lower, even if it’s not quite at bottom) after a bust, then sniffing the air and selling off stocks after a sustained boom, seems likely to net you an extra chunk of change in the long term, cumulatively speaking. True–you might not catch it at absolute bottom (you may purchase, and then see your investment decline for a while), and you might not sell at the peak (it’s so frustrating to see the selling price climb ever higher after you’ve sold), but in general, you’ll have come out ahead of those who held on through the peak, hoping for ever higher gains, and then panic sold when it fell below the price that the paid for the stock.

For those not familiar with stock stuff, this is what most investment people would consider a conservative, or low-risk strategy. It’s not gonna make you rich overnight, but over time, you should have a tidy nest egg. Since I haven’t seen a lot of evidence (and I’ll admit I haven’t looked a lot) to support the idea that other strategies are actually effective and not just a matter of those randomly generated strings that look like actual patterns, that’s what makes the most sense to me. Other options seem rather a lot games of chance, rather than calculated risks–and I prefer the latter when it comes to my savings.

At any rate, lots of overlap between the two books. Given that, the Perimeter Institute talk was more a refresher on a few of the interesting conceptual aspects of randomness, rather than a mindblowing introduction to concepts never before seen by yours truly.


Blog at
Entries and comments feeds.