E-mail a friendWilliam Poundstone
airdate February 15, 2008
Twice-nominated for the Pulitzer Prize for his science writing, William Poundstone has written 11 books, as well as for network television and major publications, including Esquire, Harper's, The Economist and The New York Times Book Review. He studied physics at MIT before beginning to write nonfiction. Poundstone's books include the Big Secrets series, a biography of Carl Sagan and Gaming the Vote, in which he assesses the obstacles confronting any attempt to change the American electoral system.
William Poundstone
Tavis: William Poundstone is a best-selling author and two-time Pulitzer Prize nominee whose previous books include "Fortune's Formula." His latest comes just in time for this election year. I bet he planned it that way (laughter). It's called "Gaming the Vote: Why Elections Aren't Fair (and What We Can Do About It)." Bill Poundstone, nice to have you on the program.
William Poundstone: Yes, nice to be with you.
Tavis: Good to see you. Let me start with the obvious question, although I'm not asking it out of any naiveté, why do you argue, though, in the book, first of all, that our elections are not fair?
Poundstone: Well, the problem is really spoilers and vote-splitting. If you look back at presidential elections since 1828, at least five of those elections have been determined by a spoiler candidate. So that basically means we've elected the wrong president eleven percent of the time. Now as I say, I mean, if you knew that an airplane had an eleven percent change of crashing, you wouldn't want to get in that plane, but that's basically what we've got with our system of elections.
Tavis: The most recent examples of that would be -
Poundstone: - of course, 2000 where Ralph Nader took enough votes away from Al Gore in the crucial state of Florida to hand the election to George W. Bush.
Tavis: For those who don't believe in the notion of spoiler, but prefer to think of it as voting for the candidate they wanted to vote for and, if it hurt or helped somebody else, why should I be concerned about that as long as I'm voting my conscience?
Poundstone: Yes. I think everyone really agrees with that. The problem is that our voting system penalizes people who do that and it really shouldn't. People should be able to vote their conscience without being penalized some way.
Tavis: Which raises the next question which is what the book is all about. How do we fix that?
Poundstone: Well, there's a lot of different ideas for that and I sort of give a whole history of this two hundred year search for the best possible system of casting ballots. But the most interesting idea in the book is something called range voting.
Now in this, the ballot is essentially like a report card and you get to play teacher and sort of grade the candidates on the scale of zero to ten. So the way it works, your favorite candidate gets the top score, a ten. Your least favorite candidate gets the bottom score, a zero. And then you rate the other candidates on that scale.
The reason it's interesting, there have been elaborate computer studies in the past few years, particularly by a mathematician known as Warren D. Smith. They make a pretty convincing case that range voting is actually the fairest of all the dozens of voting methods that people have devised.
Tavis: So to your analogy of report card, it's been a while since I've had one of these, but on the report card, it usually had either a subject or a number of other things and the teacher would go down the report card and give me a grade in specific categories. How does this report card work? What's it look like?
Poundstone: Okay. Instead of the subjects, you've got the candidates. Like if this was the Republican primary things, you'd have McCain, Huckabee, Ron Paul and so forth. You'd be able to rate each candidate exactly the way you feel about that candidate.
Tavis: Scale of one to ten.
Poundstone: Yeah.
Tavis: But just overall. There are no various categories. It's not on defense, it's not on health, it's not - okay. Just a number.
Poundstone: Yes. Just on that candidate.
Tavis: Okay.
Poundstone: And, again, you're using the whole range, so you're giving the top score to your favorite, the bottom score to your least favorite and rating the other ones somewhere in between.
Tavis: Okay. Yeah, it's been a while since I've been in school, but as I recall from my reports cards and the various classes I sat through, there would be sometimes any number of students in the class who would get the same grade. I was always in the C or D category (laughter). There were a few of us down there.
But even at the top level, there were a couple students in the class who would all get A-pluses or all get As. So what happens with this, if two candidates end up with the same grade when all is said and done, or is that absolutely improbable?
Poundstone: No, no. That's possible and it's good. Look at the situation like the 1912 presidential election. You had two Republicans, Teddy Roosevelt and William Howard Taft, running against each other. That turned out to be disastrous for the Republicans because it split their vote and it led to the election of Woodrow Wilson, a Democrat, who probably would not have won otherwise.
Well, if they'd been using a range ballot, some who liked both Taft and Roosevelt could have given the top score to both of them, and that's okay.
Tavis: But when it's all said and done, it's pretty improbable that two people could end up at the end of it with the same score.
Poundstone: It just depends on your feelings. I mean, in any race, like there are people who are right on the fence between Obama and Clinton. And if that's the way they honestly feel, they should give the same top score to both of them.
Tavis: So let me ask you honestly, then, what the challenges are to using this range voting as the best option that we have available to us. There's got to be some down side to it as well, though.
Poundstone: Yeah. That's part of the story I tell in the book. It is a very controversial system. There are a lot of people who think it's not a good idea for one reason or another. Basically it comes down to they're not sure whether the theory is going to translate into, you know, the actual practice.
One concern is maybe it's just too complicated because I think we can agree that there's a lot of voters out there who couldn't set the time on their microwave oven (laughter). You know, you have to really take that into account.
So what they're doing, it's very important to do studies, you know, showing how people are actually going to fill out these ballots. One example, there was a very well-designed study in 2002 in the French presidential election.
They had exiting voters fill out essentially approval ballots which is the simplest form of range voting on the same slate of candidates that they'd just voted on for real. They found that they filled them out, you know, just the way people thought they should and, perhaps even more important, the voters said they found this system easy to understand.
Another argument is the way that range voting has actually become very popular on the internet. On things like YouTube and Amazon, you can rate the videos or books or something and it'll give you the average rating, which is essentially the range voting idea.
In the book, I talked with James Hong who's the co-founder of a website known as HotorNot. I don't know if you're familiar with that, but the basic idea is they've got college students looking at these pictures of attractive women and you sort of rate them on a scale of one to ten to see which gets the highest average rating.
Well, Hong said he actually tried a number of different voting methods for that site and he found that range voting was actually the easiest one on the users. When you think about it, their voters are basically, you know, drunken frat boys sitting in their room at two a.m. So if they can handle this, I think the rest of us probably can too.
Tavis: (Laughter) Maybe so. The question that comes to mind now is how seriously damaged our system of voting is now relative to the need to fix the system, whether it's with range voting or something else.
Poundstone: Yeah. I think one area where all the experts are agreed is that the plurality vote is really the worst system of all that anyone has actually invented. One area where it's particularly bad is in our political primaries because there you don't sort of artificially have two choices yet. In fact, you often have four or five strong candidates in a race.
If you look at what happened to the Republicans, I mean, McCain did very well because he was the only moderate after Giuliani dropped out, but you still had, until recently, Mitt Romney, Huckabee and Ron Paul all sort of carving up the conservative vote.
So this was good for McCain, but not so good for the others. Then when Romney dropped out, you know, suddenly Huckabee had this big boost. But does that actually mean that people suddenly decided they liked Huckabee? Well, probably not. It's probably just that - you know, it's sort of a mathematical fiction because he wasn't splitting the vote.
We all agree that momentum is very important in deciding on a nominee, so it's important to get rid of this sort of artificial momentum you get by this vote-splitting. So a lot of people think range voting would be a particularly good choice for our presidential primaries.
Tavis: I wonder whether or not - on balance, it sounds like not, but I want to just explore a little more. There's always this argument always lurking of partisanship and whether or not one party, for whatever reason or reasons, benefits more from any particular change in how we adjust our electoral system.
There's always an argument against same-day voting. Republicans tend to be against same-day voting. People are against voting on Saturdays as opposed to - there are all kind of reasons for people, as you well know. I raise all that to ask whether or not there's any particular evidence that one party or the other, given how races tend to work out, benefit more from range voting?
Poundstone: No. I agree that's a very important issue and it has been discussed. There was a study actually done in the 2004 presidential election where they asked exiting voters to fill out range ballots on, you know, Bush, Kerry and the other minor candidates. They found there was no statistical significant difference between the way that the Kerry voters voted and the Bush voters voted, so that's a pretty good evidence that there's not a big partisan difference there.
Tavis: We really just kind of scratched the surface on this, but given that we are in a very controversial and, in some ways, heated election season sort of on the Democratic side as we speak, this book might be good reading for you. It's called "Gaming the Vote: Why Elections Aren't Fair (and What We Can Do About It)." It's written by the best-selling author of a number of books, William Poundstone. Mr. Poundstone, nice to have you on the program.
Poundstone: Yes. Good to be here.
Tavis: Glad to have you here.
Back to February 15, 2008