As ever the article starts at the end of the process not the beginning.
The job of politics is to bring about consensus. If you don't get consensus prior to a vote, then all of the candidates that could have done a deal ahead of the election suffer.
Under a first past the post system, the job of the candidates is to horse trade so that there are only two candidates at the election.
First past the post works because it is simple, understandable and it forces politicians to do the deals ahead of the election - or suffer the consequences of being ignored entirely.
Politicians and researchers hate it, because they want to trade for their favourite hobby horses after an election when they no longer have to suffer the scrutiny of the electorate and can use minority positions to drive forward change that would never get anywhere in a consensus system.
About the only improvement to a FPTP system you need is a 'none of the above' option, so that everybody on the list can be rejected.
> First past the post works because it is simple, understandable and it forces politicians to do the deals ahead of the election - or suffer the consequences of being ignored entirely.
Let's change this slightly:
> First past the post works because it is simple, understandable and it forces politicians to do the deals before voters have expressed their will - or suffer the consequences of being ignored entirely.
If I elect politicians who are likely to trade away list A then that indicates I either don't value list A so highly after all, or that the politicians are liars. I don't see how your system solves the latter problem - either way the solution is to not vote for politicians after they have proven themselves liars.
It seems to me that in FPTP you end up with both parties trying to calculate the absolute minimum they have to promise to their base in order to get just enough of their vote to win the election by 50%+1. And, after the election, trying to calculate what is the minimum amount of promises they have to keep, without depressing voter enthusiasm so much that they lose the next election. (The Democrats seem especially bad about this, and I say this as someone who votes Democratic and volunteers for Democratic causes and candidates.)
In others words in FPTP you have politicians trying to figure what is the bare minimum they have to do in order to not get fired. These are perverse incentives, and they do not tend to attract the sort of person I want to vote for.
Ranked choice voting seems to go a long way toward addressing this: I simply rank candidates in the order that they match my preferences, perhaps also accounting for their perceived trustworthiness as well. And, more or less, the candidate who best matches the aggregated preferences of the voters will win.
If anything, you make a great point that "playing politics" is a skill in its own rite, is worth mastering, and it is (dependent on the system) productive in bringing about people's wishes.
The job of politics is to bring about consensus. If you don't get consensus prior to a vote, then all of the candidates that could have done a deal ahead of the election suffer.
Under a first past the post system, the job of the candidates is to horse trade so that there are only two candidates at the election.
First past the post works because it is simple, understandable and it forces politicians to do the deals ahead of the election - or suffer the consequences of being ignored entirely.
Politicians and researchers hate it, because they want to trade for their favourite hobby horses after an election when they no longer have to suffer the scrutiny of the electorate and can use minority positions to drive forward change that would never get anywhere in a consensus system.
About the only improvement to a FPTP system you need is a 'none of the above' option, so that everybody on the list can be rejected.