So yeah, brackets.Hargrimm said:How about a ranked system? What about more than one round with elimination?
Brackets are fine for tiered competitions when the order of the contest matters.
However, when looking for a population statistic (I'm treating those who voted as inclusive, since this is popularity, though technically this should be a sample statistic since I don't know if everyone voted), they introduce more error than raw popular sampling.
Sure, you get results, but you miss out on other data.
Well, since the only logical criteria for 'GOTY' here is popular opinion, in the context of "most popular game", that fits perfectly.It is weighted, just not in an obvious way. Most people aren't going to vote for a game they haven't played, so the poll is skewed in favour of the most played games, which tend to be AAA games.
If AAAs are more likely to be played, they are more likely to succeed.
Personally, I'd have loved to see an indie game win, but that isn't what happened, now is it?
There isn't much you can do here to fix that.
Yup.PLaying all or even most of the games on that list would cost more money than most people are willing or able to spend on video games in a single year.
And this same problem you present is amplified via bracketing.
Why?
Well, what does the person do if they come across two games they didn't play? (Assuming they're being honest)
Or one game they didn't play vs one they didn't like?
Not vote? Well, now you've just introduced an element of exclusion into a popular vote, and that's a source of error.
That the population votes for the game they liked the most for that year, obviously.First off, what does "the more realistic sample"[sic] even mean in this context? What measure is "realistic" in this context?
Unless you're trying to objectify the relative qualities of titles, in which case a popular vote is not the way to do it, at all.
That's an arbitrary limit (or arbitrary significance), and not an objectively defined one.Secondly, what this sample tells us is that there is no one game that could honestly be called GOTY, since even the one with the most votes got less than one out of seven.
What threshold do you consider?
Greater than 50%? That's just majority rules, and it it isn't any more conclusive, or less arbitrary than any other threshold, despite its common usage.
In a popular vote system, there are only two fields we can always assume (the vote cannot exist without): Maximum, and minimum.
In the absence of arbitrary thresholds, you can make simple, decisive conclusions based on those alone.
Just by showing that ME3 won with such a small % of the vote shows that we have a diverse selection of preference.
That's useful information, and information I would not be able to definitively tell through the use of brackets, because I don't know how many times each person voted within those brackets (vote one here, skip another there because I didn't play those games..vote the next bracket in).