From Scientific American


by William Poundstone. Hill and Wang, 2008

This book will not reassure you: the U.S. has the worst of all possible voting systems. Known as plurality voting, it awards the prize to the candidate who gets the most votes among several contenders. The problem is vote splitting, the phenomenon in which two candidates split the support of like-minded voters and put someone who is not the most popular choice in office. Most of us will flash back to Ralph Nader in 2000. But the author reminds us of other cases—William Howard Taft and Teddy Roosevelt, for example, who split the Repub­lican vote in 1912, leaving Democrat Wood­row Wilson to win. By Poundstone’s calculation, in 45 presidential elections since 1828, at least five have been won by the second most popular candidate. “That’s over an 11 percent rate of catastrophic failure,” he writes. “Were the plurality vote a car or an airliner, it would be recognized for what it is—a defective consumer product, unsafe at any speed.”

Often such vote-splitting “spoilers,” the author points out, are financed by those who oppose their politics: in 2004, for example, Republicans paid for Nader signature drives, but it’s a sad bipartisan practice. Poundstone, a writer who is fascinated with how scientific ideas—those of mathematics, in this case—play out in everyday life, recommends something called range voting as the least unfair of all voting methods. In this system, voters assign rankings to candidates, and the one with the most points wins. If the 2000 election had used range voting, for example, instead of having to cast a single vote for Al Gore, George W. Bush or Nader, voters could have rated each candidate on a scale of one to five, and the candidate with the highest ranking would have won.

Click the title to find the book on Amazon.