You can’t always get what you want

I’d like to start this post with a quick detour into voting systems. And by that, I mean, the systems we use to count our votes to determine who wins. There is an _amazing_ book that should be required reading for all voters called Gaming The Vote. This book goes into a bit of history and demonstrates why the current way we count votes is completely terrible and often leads to results that do not reflect the will of the voters. The central thesis is that when you have more than two candidates in an election, the plurality system (she who gets the most votes wins) totally breaks down in many un-obvious ways.

One of those ways is that candidates can get elected with a small percentage of the vote total. Think about that one factor. There are 4 candidates and you love A, B is ok, C is eh but you hate D. In fact, the D candidate is pretty different than A-C in what they believe and A-C are pretty interchangeable to a large group of voters. (eg. D voters hate candidates A-C and A-C voters hate D).

The totals end up 30% A, 19% B, 17% C and 34% D. Is it right that D wins with only 34% approval? If the people who voted for B and C knew D would win and A had the best shot they would have gladly voted for A and thus A would have clearly won. If you think about it as A-C vs D (which is kind of is when you consider the platforms of the candidates) it’s clear.

The above problem not only shows you how shitty the results can differ from the real will of the people they also demonstrate another problem with plurality voting: strategic voting. People start to cloud their decisions by thinking about who actually can win the election rather than who they want to win the election. You can probably see a ton of examples in the past where this was an issue and imagine a ton more where it will be an issue.

I won’t get deeper into this, but this kind of stuff, to me, is extremely fascinating and the subject of many, many studies. What I will tell you is that while the studies often suggest differing solutions _all_ of the studies say that our current system is the worst. Two of the most popular systems that I’ve heard about are Approval Voting and Ranked Choice.

Approval Voting is beautiful in its simplicity. It requires (almost) no changes to machines or the math. When you go to vote, you vote for all of the candidates you approve of. So in the example above if you were an A, B or C voter, you’d vote for all 3. If you were a D voter, you’d vote for D alone. At the end, whoever gets the most votes will be the winner. Just take a second to think about how incredible this system is. Its so clean and elegant, it’s the system (last I checked) that the International Society of Mathematicians uses to pick their officers.

The other system that I like a little less is called Ranked Choice. The reason I don’t like it as much is because it’s a bit more complicated and I think that may turn off some voters and voter turn out is low enough. However, it’s gaining some traction (I think it was used in Maine in the last election) so I am totally for it. In this system you rank your choices when you vote (eg. A, C, B, D) and they use a bit of math to compute who wins based on runoffs: If no one has 51%, they drop the lowest vote getter and reallocate their votes based on their rankings and then recompute then repeat if necessary. This system again is _way_ more representative of the will of the people. I think this Radiolab episode (which I heard 2-3 months ago) talks about an election in Ireland that uses this system.

Ok. So where am I going with all of this?

The way we choose our presidential nominee is terribly unfair and will not reflect the will of the people.

This is not only based on the voting system (see above) but also on the fact that we do primaries on different dates and also a bit on the fact that our elections are not publicly funded (maybe we dig into this another time).

So, assuming that we had our primaries on the same day, what would happen is that the field would be incorrectly split up based (mostly) on ideological leanings, on the ways the candidates are different and also the same.

In very broad strokes:

by Economic beliefs : (Bernie and Warren) vs. (Kamala and Gillibrand) vs. Booker vs. Klobuchar

which would mean that the election would be between Cory and Amy since the others would split each others votes.

by Sex: (Bernie and Booker) vs. (Kamala, Gillibrand, Warren, Klobuchar)

so this would come down to Bernie v. Booker.


Of course its a lot more complicated when you come up with different venn diagrams and put them all together, but I feel that the election will be decided in these ways. This is how Trump became the nominee. He was so different than the rest of the field that they all split up the opposition to him in every primary. Will this happen if Joe Biden joins the race?

The moral to this story is that I feel that unless we change the way we vote in America, we will never get the candidate we all are happy with. And that’s sad because in every election, such a candidate does exist.

3 thoughts on “You can’t always get what you want

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s