Which two states do not have a winner take all process for president when it comes to the Electoral College?

Many who dislike the winner-take-all Electoral College argue that its bias toward small states is unfair.

That’s because each state is awarded electoral votes based on the number of representatives it has in the House, which is roughly proportionate to its population, plus the number of U.S. Senators, which is the same for all states. That means of the 538 total electoral votes, 81% are awarded by population while 19% are awarded equally.

Nate Cohn explains the circumstances where this modest bias can prove decisive:

A near Electoral College tie, as in 2000. After falling short in Florida, Al Gore lost to George W. Bush by five electoral votes, less than the net 18 votes Mr. Bush gained from small-state bias. But for perspective, that’s the only Electoral College outcome since 1876 that was within the 20 or so electoral-vote margin for the small-state bias to matter.

But this small-state bias actually had little to do with Donald Trump’s win in the 2016 election. Trump actually won seven of the 10 largest states, and Hillary Clinton won seven of the 12 smallest states. Overall, the bias towards smaller states only cost Clinton about four votes, which was not enough to change the outcome of the election.

Instead, a more important bias comes from the (mostly) winner-take-all Electoral College and how states award their votes to each candidate.

We Evolved to a Winner-Take-All Electoral College

What most interesting about our current system for selecting a president is that it’s an unintended quirk that isn’t even mentioned in the U.S. Constitution. States determine how they select their electors. In fact, for the first 13 presidential elections, states experimented with many different electoral systems.

By 1832, every state except South Carolina awarded its electors by the popular vote, although not all states followed the winner-take-all custom that emerged later. Since 1868, every state has awarded its electors in a way related to that state’s popular vote tally.

Today there are two states that do not use a winner-take-all approach: Nebraska and Maine, which split some of their electoral votes by the winner of each congressional district. They assign two votes to the plurality winner of the state’s popular vote. These two votes represent the two electoral votes they are entitled to from their U.S. Senate delegations. The other electoral votes in these states are given to the plurality winner of the popular vote in each separate U.S. House of Representatives district.

That shows that states could, if they wanted, create an electoral system that better reflected the popular vote. For instance, they could decide — as many states have already planned through an interstate compact — to award a state’s electoral votes to the winner of the national popular vote.

Of course, there are strong incentives to not move away from a winner-take-all allocation for the same reason they moved to the system in the first place: It would dilute their power in the election of the president.

The winner-take-all system came about because of partisan power. Once some states came to this conclusion, others had no choice but to follow to avoid hurting their side.

The Electoral College Is Biased Towards Larger Battlegrounds

So while the Electoral College was designed with a built-in small state bias, it isn’t nearly as big as the advantage that a winner-take-all system gives to the larger battleground states.

We saw this clearly in the 2016 presidential election. Trump won the election because his political coalition was efficiently distributed among the Midwestern battlegrounds of Wisconsin, Michigan and Pennsylvania. He won each of those states by a narrow margin, but was able to claim all of their electoral votes.