In a winner take all system only the winners win - here's the fix

How much does your vote count? That all depends on whether you voted for the winner or the loser.

According to 2012 presidential election data compiled by the Roper Center for Public Opinion Research at Cornell University, there were about 220,000,000 voters eligible for the presidential election, resulting in about 127,000,000 votes cast for one or the other of the major party candidates - or about 58% of the electorate.

The winner, Barrack Obama, received about 66,000,000 votes, or nearly 52% of the votes cast for either a republican or a democratic candidate - which, according to our system of elections, are the only votes that effectively count towards the outcome.

If we look at it a slightly different way, the President was elected by roughly 30% of the eligible voters. These figures, which vary only slightly from election to election, have become extremely familiar to Americans. This is often held up as one of the reasons Americans are generally dissatisfied by the results of any given election. "There's something wrong," people are quick to say, "with a system in which the most powerful person in the country is elected by less than 1/3 of the people."

Very likely true. However, completely peripheral to the actual problem. The actual problem has much more to do with the value and power of any individual voter's influence on the outcome of the election. One reason a large minority of voters don't vote is the perception - the accurate perception - that individual votes matter very little. Less than a drop in the bucket. This is statistically true. Let's look at it:

Each voter represents 1/220,000,000 of the total potential electoral power. This is slightly complicated by the Electoral College system - but we'll come back to that. Let's just focus on the popular vote for a moment. If we were to assume 100% participation, each vote would count 0.0000005% towards the outcome. The good news - if you do vote and feel very insignificant - is that the lower the turnout, the more power you have. Based on the 2012 statistics, your vote counted 1/127,000,000 - or 0.000001% towards the outcome, roughly double the power you would have had had everyone voted.

In a binary winner-take-all system such as our national presidential election, the only votes that matter are the winning votes. So assuming you voted for Obama in 2012, your vote represented a whopping 0.000002% of the effective electoral power - once again doubling the value of your vote. Good for you!

But, of course, the reality is your vote only counted if it was cast towards the winner. If you voted for the loser, your vote counted 0.0000000-to-infinity-% towards the outcome. Why? Because there can only be a binary result in a two party system.

Now, let's factor in the much maligned Electoral College. Here's how that works:

As you probably know, each state has electoral votes equal to their representation in Congress (total of Senators and Representatives). So if your state has 4 Congressional delegates, you have four electors. In all states, the Electors are expected (generally by state law, now) to vote according to the popular vote winner of their states (with two exceptions: Maine and Nebraska allow a split electoral vote - which may sound more fair, but doesn't in the end matter at all). As of 2012, there were therefore 538 electoral votes. The winner of the presidential election needs a majority of these to secure victory (half +1), which means 270 electoral votes are needed to be elected president. Just like with the popular vote any non-winning vote is irrelevant and has no bearing on the outcome.

Thus, 268 of the 538 votes automatically count 0.0000000-to-infinity-% towards the outcome. If you are on the losing side of the electoral college, it makes the situation no worse than being on the losing side of the popular vote.

Stated another way - each of the 270 winning electoral votes in 2012 represented 814.815 individual eligible voters - regardless of how many actually voted, and regardless of which way they voted. The effective influence of each winning electoral vote was therefore 814,815 /220,000,000 = or 0.37%, which multiplied by 270 of course yields 100%. The math can never work any other way because only winning votes count, and winning votes represent 100% of the electorate, no matter which way the electorate actually voted.

Looked at another way: the winner gains the benefit of all non-votes and does not lose anything for votes against them. All votes - for, against, and non-votes - ultimately count in the winner's favor by the time it gets to the Electoral College. Hmmm...

This little statistical anomaly is why the parties do not bother to campaign in states where the popular vote is unlikely to be close, or where voter turnout is likely to be high. Winning an Elector by more than 1 vote is a waste of time and money; whereas losing an Elector by a million votes is no worse than losing them by 1 vote (as the famous Florida hanging-chad result illustrates).

States that favor one side or the other, or where voter turnout is generally high, offer a very low chance of influencing the outcome. The parties both strongly prefer a low voter turnout in the general election because of the magnification effect of the Electoral College.

This has had a decisive effect on just how campaigns are conducted and why negative campaigning has become the predominant modus operandi of the political parties. Both parties gain a far greater likelihood of success by keeping undecided voters home than by trying to get undecided voters to vote for them. In other words: a non-vote always has a statistically more significant influence on the outcome, because it magnifies the value of winning votes. For the math geeks out there - decreasing the denominator has a much greater effect on the answer than increasing both the numerator and the denominator. As an example - an increase of 1 to both changes the equation from 3/5 to 4/6, which increases the result from 60% to 67%; while changing it from 3/5 to 3/4 moves the needle from 60% to 75%. This is geometrically enhanced by the Electoral College effect, as shown above.

Here's the real problem: In a two-party, winner-take-all system only winning votes count – however statistically insignificant they may be. Look at it this way – a vote for the losing side is effectively a vote for the winning side, because in the end it does not change the outcome in any way.

If you'd like to apply some pressure to the political parties - or to otherwise affect a positive change in the system - here are some simple state-level fixes that would require no Constitutional or Federal election law changes (which would be awfully hard to achieve, considering the parties' monopoly on Federal politics, and the fact the proposed changes are entirely aimed at decreasing the parties' power):

1 - Calculate the popular vote winner in each state over the full denominator - all eligible voters in the state. This would create incentives for both parties to encourage - rather than discourage - voter participation, eliminating their ability to manipulate the outcomes by manipulating the "denominator variable."

2 - Allocate no Electors to any candidate that does not receive a true majority of all eligible voters in the state. This would break the inherent monopoly the two parties share over the Electoral College by creating a third - "undeclared" - option for voters. The parties would need truly decisive state victories to receive any committed electoral votes, and encourage a truly national campaign. The real effect of this would be to localize the presidential election, by making more state results more meaningful. The harder it is to secure the required majority, the harder the parties will need to work to win every vote in every state. The likelihood of only two or three states becoming decisive in any given election would be greatly diminished.

3 - Instruct state Electors to abstain during the Electoral College vote where no candidate received the required majority. This would eliminate the rubber stamp effect, and ensure no inconclusive and non-representative election received an irrevocable and fundamentally artificial endorsement. The likelihood of any marginal, non-representative candidate receiving the required half+1 electoral votes would be sharply reduced.

The reality of such a system is that the Electoral College would be less likely to achieve a decisive result. This condition would persist as long as the two national parties continued to behave as they do today - and also ensure that the popular vote was more meaningful up to and including the congressional vote. Congressional races would take on a far greater significance than they currently have - because the composition of any state's congressional delegation would have a profound impact on who will become president.

There are existing Constitutional provisions for handling an indecisive Electoral College result. The election would be passed to the newly seated House of Representatives (from the most current election) in a so-called "Contingent Election." Each state delegation receives 1 vote regardless of how many seats they hold; small and large states wield equal power. The indirect - but ultimately desired - effect of this is that local politics would take on much greater power and influence over the national agenda - which, ultimately, was the intent of the Founding Fathers to start with. We would end up much closer to a bottom-up process than the current top-down process.

As we know, third parties are far more powerful on the local and congressional level than they are at the national level. This would give a stronger voice to the "outsiders" who are excluded from the two party system.

One or two election cycles so decided would shake the present non-representative system to it's core. And, at the very least, it would be a lot more fun for us, and a lot more uncomfortable for the major parties!

More importantly - let's recall what the office of President was intended to be under the Constitution. The President was intended to a national statesman, a man or woman above the regional and partisan fray; a person to act as a fair and reasoned arbiter and guardian against partisan politics. The changes proposed would make it far more difficult for a candidate with too great an association with one of the two major parties to become president. The candidates more likely to achieve a decisive result would be notably bi-partisan or even - perish the thought - non-partisan.

Interesting to think about.