<p>Note how this thread is starting to devolve into current politics. Whether or not you agree with the decision made by people via a procedure has nothing to do with the merits of the procedure. Any mention of who has been elected in the past is a red herring. You think Americans are stupid? This is exactly what is stupid about them: if one doesn’t get his way, he blames the system for allowing “idiots” (read: “people who disagree with him”) to have a say in the outcome.</p>
<p>The right way to compare different decision-making procedures is with logical criteria. You can prove whether a procedure satisfies a criterion, and you can decide unambiguously whether the criterion is desirable. Read up on some basic game theory. It might make you smarter.</p>
<p>For example, it has been stated here that major parties in addition to the top two would have to steal from the top two, splitting the vote and making it difficult to win the most votes. This is called the spoiler effect. It affects plurality, but not a single-winner voting system that is [independent</a> of clones](<a href=“http://en.wikipedia.org/wiki/Clone_independence]independent”>Independence of clones criterion - Wikipedia). The [ranked</a> pairs](<a href=“http://en.wikipedia.org/wiki/Ranked_pairs]ranked”>Ranked pairs - Wikipedia) system I mentioned earlier is independent of clones.</p>
<p>America’s two party system is the result of Duverger’s law, which is in turn an effect of the plurality election. Proportional representation is not the only way to avoid Duverger’s law. A voting system that meets the [Smith</a> criterion](<a href=“http://en.wikipedia.org/wiki/Smith_criterion]Smith”>Smith criterion - Wikipedia), e.g. ranked pairs, also avoids it.</p>
<p>Repede, note the question: “If you had the power to pass one national law right now, what would it be?” I’m not hopeful that America will adopt a national popular vote / ranked pairs, but I believe it should.</p>
<p>Recounts are only necessary when errors are likely. Errors can be eliminated by eliminating potentially faulty parts of the system (i.e. human readers and instruments that convert optical or tactile readings into information). I think the technology is easily available to digitize the process entirely.</p>
<p>As for ranked pairs’ complexity, people don’t have to understand the algorithm. All they have to know is that they rank any number of candidates on the ballot, from best to worst. They could technically just vote for their favorite candidate, or they might have a top 10.</p>
<p>LCengineerDT, I don’t think you quite understand how the electoral college works. People don’t vote for electors. They vote for candidates. (In some states, a candidate is listed next to the elector(s) who pledge to vote for the candidate.) A state government tallies the ballots cast by its own residents and uses this result to determine which people to send to the electoral college. In 48 states, all electors are told to vote for the state-wide winning candidate. In the other two states, two electors vote for the state-wide winner; for each congressional district, another elector is told to vote for the winner within that district. In the absence of faithless electors (which are discouraged universally and punished in many states), this is deterministic. I don’t know how much you know about basic math, but “deterministic” means the same input always produces the same result. There is no feasible opportunity for a middleman to decide that people voted stupidly and to contradict their collective decision. Therefore, the electoral college offers no benefit over a popular vote.</p>