MOBAFire's first Mini Guide Contest of Season 14 is here! Create or update guides for the 30 featured champions and compete for up to $200 in prizes! 🏆
Just gonna mention that my guide has 22 votes - 19 of these were friends who I asked to look at the guide, one was a 'herpa durpa dis guid sux cuz i haz better wun' troll downvote, one was an annonymous downvote and ONE was an actual legitimate impartial vote.
Also, the two troll votes took it down 6% and the two subsequent legitimate upvotes took up by 1% combined.
System is flawed at best.
Also, the two troll votes took it down 6% and the two subsequent legitimate upvotes took up by 1% combined.
System is flawed at best.
Didn't they try permanent C2V already?
If I remember right it made for ridiculously inflated scores.
If I remember right it made for ridiculously inflated scores.
That's because a downvote is worth more than an upvote - and that should ALWAYS be the case.
Imagine you're reading the newspaper. And four people were asked to review a new brand of tuna. 3 of reviews are positive and 1 is negative. The negative review will always stick to your mind the most.
"Tasted nice", "Pretty bland", "Alright taste", "Reminded me of SpringWater Tuna".
You're going to remember "pretty bland" and perhaps not buy that tuna, or maybe you'll try it once.
The negative outweighs the positive because too many people upvote and not enough people downvote to equalise high scores.
My theory anyway.
Imagine you're reading the newspaper. And four people were asked to review a new brand of tuna. 3 of reviews are positive and 1 is negative. The negative review will always stick to your mind the most.
"Tasted nice", "Pretty bland", "Alright taste", "Reminded me of SpringWater Tuna".
You're going to remember "pretty bland" and perhaps not buy that tuna, or maybe you'll try it once.
The negative outweighs the positive because too many people upvote and not enough people downvote to equalise high scores.
My theory anyway.
Okay okay.
Let me just say that I was seriously pissed when I made this thread. Now I am a bit more calm, and realize that if my guide is awesome, it will get to where it should be one way or another.
Thanks for all the serious replies here, you have made me realize I was just being a jerk.
I apologize...
Let me just say that I was seriously pissed when I made this thread. Now I am a bit more calm, and realize that if my guide is awesome, it will get to where it should be one way or another.
Thanks for all the serious replies here, you have made me realize I was just being a jerk.
I apologize...
DuffTime wrote:
ok ok plz carry me omg
i was only waiting for you to ask
Temzilla wrote:
Too hot to be icecream.
Luther3000 wrote:
He looks like a hair gel advert on legs
Toshabi wrote:
Icecreamy, with hair as slick and smooth as the ocean waves of Cocobana
http://www.mobafire.com/league-of-legends/build/how-voting-works-114387
A little outdated (+rec's still listed) but yeah.
A little outdated (+rec's still listed) but yeah.
Snowfall
<Member>
<Member>
I like C2V, and I think you should be able to customise the number of votes.
Also, about it lowering the amount of people brave enough to downvote - If they aren't brave enough to downvote, it probably means 2 things - 1. They don't have a valid reason, so they'll get bashed for trollvoting, or 2. They're wimpy.
Also, about it lowering the amount of people brave enough to downvote - If they aren't brave enough to downvote, it probably means 2 things - 1. They don't have a valid reason, so they'll get bashed for trollvoting, or 2. They're wimpy.
The system for determining rank is based on a commonly used formula that attempts to predict what the final score will be once it has a sufficient number of votes, based on a smaller amount of votes. It attempts to protect guides with limited voting information from getting ranked too high or too low while also giving newer guides with sufficient voter information a chance to compete with old guides.
Obviously this isn't perfect, how can it be, it is guessing. However we did not write the formula ourselves, and it is widely accepted as the best way to handle this. There are a few similar formulas, but in my testing they never deviate from our current formula by very much and they all have their own issues.
Reddit for example uses a formula just like ours, I have tested their formula on the site and the results are quite close. We've discussed switching over to Reddit's formula but are afraid there would be a massive amount of backlash from authors when their scores change, even if it happens to everyone and it's not a significant change. It also has its own issues that would need further systems to deal with.
So yes, you may see strange fluctuations in your score as votes come in, but no it is not "flawed", it is doing the best with what little information it has.
You may also be confused by the caching on the site. Scores do not update right away and it is unpredictable due to the timing. This can be misleading if you are watching your score like a hawk. Your guesses as to how (anonymous) votes affected your score are pretty much guaranteed to be incorrect.
Regarding C2V, this will never happen. The only thing that may ever happen is that we pull the current 20 C2V limit altogether. The feature is a disaster. I have explained why dozens of times elsewhere, if you do some digging I'm sure you'll find one of various essays I've written about why it's a disaster. Customizing C2V for each author is even worse. Now everyone is on a totally uneven playing field.
Voting is meant to be anonymous. That's the entire point. Having a committee reviewing votes completely defeats the purpose, whether it's the author or a selected panel or the general public. Forcing people to vote publicly is a huge mistake and leads to nothing but retaliation, hate, anger, censorship, unrealistic scores, and yes, trolling.
If any system is needed, it's one that drives MORE votes to MORE guides. Not one that filters, censors, blocks, deletes and manipulates them.
Obviously this isn't perfect, how can it be, it is guessing. However we did not write the formula ourselves, and it is widely accepted as the best way to handle this. There are a few similar formulas, but in my testing they never deviate from our current formula by very much and they all have their own issues.
Reddit for example uses a formula just like ours, I have tested their formula on the site and the results are quite close. We've discussed switching over to Reddit's formula but are afraid there would be a massive amount of backlash from authors when their scores change, even if it happens to everyone and it's not a significant change. It also has its own issues that would need further systems to deal with.
So yes, you may see strange fluctuations in your score as votes come in, but no it is not "flawed", it is doing the best with what little information it has.
You may also be confused by the caching on the site. Scores do not update right away and it is unpredictable due to the timing. This can be misleading if you are watching your score like a hawk. Your guesses as to how (anonymous) votes affected your score are pretty much guaranteed to be incorrect.
Regarding C2V, this will never happen. The only thing that may ever happen is that we pull the current 20 C2V limit altogether. The feature is a disaster. I have explained why dozens of times elsewhere, if you do some digging I'm sure you'll find one of various essays I've written about why it's a disaster. Customizing C2V for each author is even worse. Now everyone is on a totally uneven playing field.
Voting is meant to be anonymous. That's the entire point. Having a committee reviewing votes completely defeats the purpose, whether it's the author or a selected panel or the general public. Forcing people to vote publicly is a huge mistake and leads to nothing but retaliation, hate, anger, censorship, unrealistic scores, and yes, trolling.
If any system is needed, it's one that drives MORE votes to MORE guides. Not one that filters, censors, blocks, deletes and manipulates them.
^ I agree.
One thing I thought of might be relaxing the rules on guide advertisment just a little bit - for example, allowing each person to post a single thread on their guide in an appropriate forum. I don't know how practical that is though.
As someone who recently published my first guide, it is impossible to get any feedback from anyone except friends.
Perhaps expanding the various existing reveiw threads into something more official would help too - I don't know if the scout forums already have something like this.
One thing I thought of might be relaxing the rules on guide advertisment just a little bit - for example, allowing each person to post a single thread on their guide in an appropriate forum. I don't know how practical that is though.
As someone who recently published my first guide, it is impossible to get any feedback from anyone except friends.
Perhaps expanding the various existing reveiw threads into something more official would help too - I don't know if the scout forums already have something like this.
You need to log in before commenting.
That's not it.
It's the reason that once it's there, people new to Mobafire will use that one since it's at the top, do well and be all like "WOW THAT GUIDE HELPED ME IT MUST BE THAT GUIDE" and upvote it.
Same basic principle.
People don't want to look any longer then they have to for 'The Guide.' *derp*