On Sunday primaries are happening in Argentina, so time for some electionposting!
"Primaries" is what they're called, anyway, but it's a misnomer. Nominally, parties could have multiple pre-candidates for any given position being elected, and if they did, people would choose one of those pre-candidates and vote (pre-vote?) for them (only one per position. you can vote in party A's primary for president and party B's primary for governor, but you can't vote for both party A's and party B's presidential primary). And if you got more votes than every other pre-candidate for the same position in your party, then you get to be the candidate and run in the general election in October. That is, as they say, the theory.
In practice, parties prefer to choose their candidates via internal processes. Depending on your jurisdiction either none or almost none of the positions you can vote for will have multiple pre-candidates for any position. As a secondary function, parties that did not get more than 1.5% of the votes across all pre-candidates do not get to participate in the general election at all (per position. you can get enough votes to participate in the governor election while failing out of the presidential, etc.)
Given that nobody bothers to actually propose multiple pre-candidates and so the primaries accomplish mostly none of their purpose (except filtering out a few small parties which would not have changed the outcome of the general election), surely we could get rid of them? The system was tried, it has shown itself to be an expensive waste of money and time (remember, Argentina has mandatory voting. You can't say 'eh this is pointless' and not show up*), let's call it a lesson learned.
Well, somehow, that has not happened. At a guess, and with all the relevant caveats for stipulating the real reason for things, because the way people vote in the primaries carries a lot of information about how they will vote in the general†, the primary is in actual fact just a very well-funded electoral survey with really big sample size. Given that the data from that survey is useful for the last few months of politicking before the real election, and given that the people who could in principle say "hey, this was a bad idea" and stop them are also the primary beneficiaries (heh) of this waste of money, we continue to do it.
*Well, you probably can. The penalties for not voting are not that onerous, actually. There's a fine of between 50 and 500 ars. On the low end that is like a little more than the two bus tickets you had to pay for to go vote and back if your polling place is not within walking distance. On the high end, you're still in range of 'money I would be willing to pay to save myself a few hours of waiting in line' (which is a not-unlikely outcome of voting day). Maybe not, depending, but lots of people have a higher marginal value of time than me. Oh, or you can also ignore the fine and if you don't want to do some specific kinds of interacting with the government for the next year, this has no effect.
†In theory this need not be the case. People could observe that, if your goal is 'make it more likely the candidate I like the most/hate the least wins', your vote has zero effect past the point they secure 1.5% of the vote. If this is not in question, and for most people it isn't, then you could blank your vote or vote in the primary of the odd party who actually has precandidates or make sure a tiny party who might not get the 1.5% gets it. Or, heck, roll an appropriately sized die. But in practice people do in fact vote for the pre-candidate they will vote for in the general.