MattMaroon.com

One thing that never ceases to amaze me is the love we humans have for cherry-picking data to come to conclusions. A fine example is this Malcom Gladwell article in which, in an attempt to prove the hypothesis that SUVs are unsafe due to poor handling, he runs a Chevrolet TrailBlazer and a Porsche Boxster Convertible through the Consumer Reports test track. Unsurprisingly the $50,000 sports car (designed mainly for driving around Consumer Reports’ test track) outperformed the SUV, which costs about half as much and was designed for schlepping kids to and from baseball practice. While you’d have to be borderline retarded to consider a test-track comparison of a Boxster and a TrailBlazer proof or even anecdotal evidence of anything, Gladwell does so with glee.

More recently, you see this sort of stuff cropping up in financial advice. It’s easy to take previous peaks and current valleys and say things like “If you’d invested in the stock market in May 1999, today you’d be down 20%” or “if you bought a home 7 years ago, it would right now be worth what you paid for it then.” (Nobody mentions that between March of 2003 and October 2007 the market went up 80%, or that it nearly quadrupled just during the Clinton administration, or the home valuation increase between sometime last year and any period before that.)

People see that and say “Holy Shit, I need to dump all of my money into CDs and go back to renting”. That’s stupid. Both of the above are proof only that those two markets are subject to variance, and hell, even Wikipedia could have told you that. The fact is that even with current woes, our stock market has returned north of 10% annually since the Great Depression, and real estate has gone up around 1%.

Real estate may sound less impressive, but it’s probably the better investment, at least up to a certain point. It may only gain 1% per year on average, but that’s 1% of the value of the entire property, while thanks to mortgages you’re only putting down 10% to begin with. So it’s really a gain of 10% on the money invested. With your primary residence at least, the tax deduction on the interest makes buying not much more expensive than renting in many areas of the country in the short term. When you factor in the fact that rents increase periodically while a fixed-rate mortgage never does, over a period of, say, 10 years, it’s probably one of the best investments you can make.

On stocks, Warren Buffet says:

Over the long term, the stock market news will be good. In the 20th century, the United States endured two world wars and other traumatic and expensive military conflicts; the Depression; a dozen or so recessions and financial panics; oil shocks; a flu epidemic; and the resignation of a disgraced president. Yet the Dow rose from 66 to 11,497.

The stock market, too, will always remain a better vehicle for long-term savings than CDs. The reason is simple supply and demand. People want lower-risk investments more than higher ones, so they’re willing to accept lower returns for them. If CDs were paying more than the theoretical yield on stocks, money would flow out of the markets and into the banks, and stock market returns would go up while banks lower rates on CDs. If those, treasuries, and other forms of low-risk investments ever pay more than the market over an extended period, it will only be because all of those cell-phone waves flipped some sort of crazy switch in all of our brains.

Of course, the stock market is high-variance (as that cherry-picked data shows) so you shouldn’t stick any money in there that you’re going to need in the next 10 years. Also, as you age, you should shift your newer investments more toward the safer instruments, and at some point when you’re within spitting distance of retirement and the market is doing alright you should move it all that way. But if you’re my age and you’re sticking it all in a money market because the Dow is even over the last 10 years, you’re a victim of cherry-picked data, and a buffoon.

I think the problem is that humans don’t often question other people’s logic. When we argue, we often do so merely by showing different data and trying to draw opposite conclusions from it. McCain says Obama plans to raise taxes (data) and that raising taxes during tough economic times is bad because Herbert Hoover did it (logic) therefore Obama’s tax plan is bad for the economy. Obama doesn’t come back with any arguments that the logic (that raising taxes must be bad for the economy) is flawed or at least not universally applicable, instead he just shows different data, namely that 95% of people will get a tax cut.

Gladwell gives you some data (that a TrailBlazer handles worse than a Boxster), applies some logic (that therefore, SUVs in general must handle worse than non-SUVs) and voila, the conclusion that SUVs are less safe.

It seems that in public discourse, he who makes the first logical assertion based on any data at all has the high ground. Unless the logic is absolutely ludicrous (such as the notion that being able to see Russia from your state gives you foreign policy experience, and even that is accepted by about 40% of the population) your opponent has no chance but to try to find opposing data.

In the real world though, where logic isn’t always as sound as it may appear at first glance, data is a loaded weapon. If you know what you’re doing with it, it can be a lifesaver, but if you don’t there’s a good chance you’ll shoot yourself in the foot.

This entry was posted on October 17, 2008 at 6:50 am and is filed under Uncategorized . You can follow any responses to this entry through the RSS 2.0 feed Both comments and pings are currently closed.