Since living in Canada, I’ve had many occasions to explain to Canadians just how screwy the Electoral College used in US presidential elections is. The thing that folks frequently don’t know is that a state’s electoral vote count is not simply proportional to the state's population, but rather skewed to give more votes to the smallest population states than they would otherwise have, by *de facto* taking some from the highest population states. This is enacted by each state (and DC, yes yes) having a vote per Representative they would send to the House, plus 2 (“for the senators” you might say). That +2 is trivial when you’re California (it’s about a 4% increase in their total number of EC votes), but huge when you’re Wyoming (it’s a 200% increase in their total number of EC votes). So you can do some quick math,[^0] and get the percentage of the US population in a given state, and the percentage of EC votes assigned to that state, and find the ratio between them, and get just how skewed it is: a vote in California or Texas is 85% power, while a vote in Wyoming is 326% power. [^0]: You can [see the math here](https://docs.google.com/spreadsheets/d/1cTjFG1DZXSmkGKdYx8mbgR7eJMQC76PhqQqegzLg6kY/edit?usp=sharing). (We’re leaving aside for the moment “how likely the state is to go D or R”, but we will talk about that later.) So this is the baseline. But there’s a factor I never hear people talk about, and that is *voter turnout*. The variability between states’ turnout is tremendous, and I think it should be pretty obvious that if you are the only person who votes in your state, and you get all the EC votes assigned the way you want, that’s more powerful than if everyone in your state votes, and the EC votes go the way the majority decides—you won’t have been as decisive a factor in your state’s Electoral College choice! So what if we take the EC vote power ratio above, and combine it with turnout? The power of a vote in a state with low turnout is increased, while with high turnout, it is decreased. We don’t, as of the time of this writing, have comprehensive 2024 turnout numbers, and 2020 is admittedly a strange year (the pandemic made many states make absentee/mail-in voting more accessible, and any increase in voting accessibility increases turnout[^4]), but it was also a census year, so weirdness aside, I’ll use it for my math here: [^4]: Did you know that 2020 is the first year, from 1976 to now, where the plurality candidate outperformed “no vote” among eligible voters? [Here’s a nice chart.](https://cosocial.ca/@EricCarroll/113468181781695062) And this increase in voter turnout is likely [because of the pandemic making governments decide to make mail-in easier](https://www.pewresearch.org/short-reads/2020/10/13/mail-in-voting-became-much-more-common-in-2020-primaries-as-covid-19-spread/). The top five most powerful voting regions are: | State | EC-T power | EC-T power normalized | | -------------------- | ---------- | --------------------- | | Alaska | 372% | 1.87 | | North Dakota | 373% | 1.88 | | Vermont | 393% | 2.10 | | District of Columbia | 424% | 2.45 | | Wyoming | 504% | 3.34 | And the bottom five are: | State | EC-T power | EC-T power normalized | | ------------ | ---------- | --------------------- | | Florida | 121% | -0.91 | | California | 124% | -0.87 | | New Jersey | 125% | -0.87 | | Michigan | 126% | -0.86 | | Pennsylvania | 128% | -0.83 | Because of how the math works (no state can have more than 100% turnout), even the least powerful state’s power will be above 100%, but we can normalize the range of power, and that’s what the second column is: how far above or below the mean vote power your vote is in that state. So a Floridian in 2020, factoring in the distortion of the Electoral College and of Florida’s turnout numbers, is almost 1 unit[^1] below the average state’s voting power. A Wyomingite, on the other hand, has a vote that is over three units above the average state’s voting power (which is _a lot_.) [^1]: Standard deviation. The units are standard deviations. If that’s not familiar to you, don’t worry about it, just treat it as a “unit of farness from the mean”. As you can see from the top and bottom states here, this isn’t a particularly partisan problem, or analysis. Everyone gets distorted, more or less. So I guess you could say “who cares, it comes out in the wash”, but I don’t agree. When the alternative is simpler and more fair, you need a better reason than “it washes out” to keep doing something convoluted and unfair.[^2] [^2]: And, first off, it’s not clear that it _reliably_ washes out, even if it does sometimes. And second, I’ll grant “because it’s in the Constitution and is therefore effectively unchangeable” is a good reason to keep doing it, but a bad reason to not talk about alternatives. So who benefits from low turnout? Whoever _does_ turn out. If your state usually has very low turnout, and you think “why vote, we’re gonna go X anyway” (I’m looking at Oklahoma, Arkansas, Hawaii, West Virginia, Tennessee in the 2020 election, none of whom passed 60% turnout), you are effectively giving your vote to the few people in your state who *do*. We act on stories—we are story-driven beings—and “my state’s a lost cause” or “my state’s a safe bet” are stories that can, through the inaction they drive, become self-fulfilling. Now, none of this is a path to political change; our systems in the US are incredibly calcified and unwilling to change, and I recognize that “here’s a good idea” or “this thing is unfair” do not, themselves, create change. But I hope that this has made you at least think about what’s bad about our current system. The more people who see this system as bad, the more possible change becomes. Given my druthers, I’d not just advocate for a national popular vote for president, but also [a *ranked choice* vote](https://en.wikipedia.org/wiki/Instant-runoff_voting). But that’s a discussion for another day. --- An addendum: I know that the president of elected by the states, not the citizens. I’ve read enough founding documents. I still think that’s absurd and unfair with our country as it is now. Another thing: let’s talk about “how much your vote matters” when it’s all boiled down to “50%+1 means that the other 50%-1 get zero percent of what they voted for”.