Top Ranked Games

ID:20979
 
Oct 15 2006, 2:59 pm
I see a lot of discussion going around BYOND recently about how Top Ranked Games are not accurate. This is probably true for the most part if you think of it as a true Ranking system. I do think that the system does a really good job of showing what games are most popular from a BYOND member standpoint.

Maybe if the name "Top Ranked Games" were changed to something such as "Most Popular Games" or "Favorite Games" then it might not be so heavily debated as being an inaccurate "ranking system".

That's what I think about the whole thing anyway. :)
Oct 15 2006, 3:08 pm
The bribes throw it off a bit sadly, but other than that it seems to work well for a popularity thing.
Oct 15 2006, 3:13 pm
"top ranked games" sounds fine for me, seeming as it shows the games that have been ranked the highest. Popularity would even be a misleading name, what if the most popular game on BYOND didn't have anyone with memberships playing it (however unlikely that sounds)?
the whole ranking thing never really bothered me, however, so whatever.
Oct 15 2006, 3:21 pm
Most of the bribes involve in-game bonuses. In theory you'd have to like a game a fair bit in order to take a bribe made up of things inside the game.
For example Silk 'bribed' everyone by saying that if AA gets to number one there will be a open test. However if people like the game enough that they're willing to vote just to play it then the votes it gets can't really be considered tainted.
Oct 15 2006, 5:10 pm
Ranking is horribly biased. For one, only BYOND members can add a game to their rankings (we are told because only BYOND members have favroites, though that can be implemented to non-BYOND Members). This means that only a small percentage (as we actually are the minority) can add games to their favorites.

If everyone could vote, top ranked games would be different. I cannot stress how many more votes my game (Hurricane) would get if everyone else who plays it that's not a BYOND Member (only a hand full who've played it a lot are BYOND members).

That's one reason why ranked games aren't accurate. But as is, it is what's most popular for BYOND Members, not the community as a whole. Sure, we may like it, but what about the other 75% (guessing, it's around here) of BYOND goers? Just not very accurate in its current state.
Oct 15 2006, 5:30 pm
There's no real need for them to accurately reflect the favorite games of the entire BYOND community. It's not like there is anything riding on game rankings.
Provided that it show off some entertaining games it's doing it's job.

But as is, it is what's most popular for BYOND Members, not the community as a whole.

I don't mean to sound like the rest of the community is unimportant, but that's how the system is meant to work. The system is designed to show off what BYOND Members like. So in that respect it's 100% accurate.


but what about the other 75%

I believe it's closer to 95%, with 15% being the goal.
Oct 16 2006, 12:26 am
DarkView: Have you ever taken Statistics? Better yet, college level statistics? In it you learn about the forms of bias' and the such. This particular bias, where only BYOND Members are aloud to vote and not the entire population, is called selection bias. Essentially, it's when a group not representing the population entirely is used instead of an all around group, such as the entire population and a random selection of each (this wouldn't work for BYOND).

BYOND Member's are generally those who've been around on BYOND for some time, very few of the well known people are not BYOND Member's, it's usually the lesser known members or those who (for whatever financial reason) cannot afford to be a Member. Those "lesser" known people are usually the newer members of BYOND (when compared to most members here) and thus have a different viewpoint of games than veterans like ourselves.

This altogether is the reason why only using BYOND Members is statistically incorrect and a huge bias. If further explanation is needed, simply ask.

As a side note, thanks for correcting my percantage of how many BYOND members compared to non-BYOND Members, I was basically guessing.
Oct 16 2006, 2:14 am
My problem here isn't an understanding of statistics. As far as statistics go this is barely even highschool level stuff. That said your mistake has nothing to do with statistics. You've just got the objective of the BYOND Ranks wrong.

it's when a group not representing the population entirely is used instead of an all around group

The BYOND Ranks aren't meant to reflect the most popular games on BYOND. The BYOND Ranks reflect which games the BYOND Members rank as their favorites.

If it was intended to be a list of which games BYOND users (reguardless of membership status) consider the best then you'd be right. However the current system would still be the most accurate way of figuring that out. If you let any key vote the valid votes gained would be absolutely destroyed by the invalid votes generated by people with hundreds of alternate keys.
Oct 16 2006, 2:24 am
DarkView: Have you ever taken Statistics? Better yet, college level statistics? In it you learn about the forms of bias' and the such.

That's not college level statistics, it's just common sense.
Oct 16 2006, 8:30 am
Kunark wrote:
DarkView: Have you ever taken Statistics? Better yet, college level statistics? In it you learn about the forms of bias' and the such.

That's not college level statistics, it's just common sense.

True, statistics (as yes, I did say normal statistics, not just college level which can be a hishschool coarse for AP) simply makes use of the bias and places it in to perspective more. Yes most people can see the bias, it's just more easily explained through statistics.

DarkView: When one sees "Byond Rank:" one would tend to think the general ranking of the game for all byond goers, not just the members, that was where I was getting at, though it was probably meant for only the BYOND Member's viewpoints instead of the census, but I was basing what I said off of what the site says, not by interpretation.
Oct 16 2006, 3:26 pm
Well im sorry to say this but if the people who arent members, really care about the whole 'ranking' thing so much, go out and buy a membership to support the community. They are not expensive and last a full year. Non members opinions do matter, but they shouldn't affect the so called "rankings" if they dont buy membership to better support byond. "rankings" are a privelege to those who buy memberships. Just like in any club or orginization, being a member has its perks.
Oct 17 2006, 5:47 pm
though it was probably meant for only the BYOND Member's viewpoints instead of the census

There's no probably about it. I'm not saying that because it's how I think it works and I'm always right, I'm saying that because I was close enough to the development process and that's what it was intended to be (brainstorming, not the building =P).
Basically rankings is a way for BYOND Members to see what games other BYOND Members like. However non-Members may find this information useful/interesting, so the rank list is presented on the main site along with the rest of the hub.
Dec 22 2006, 2:13 pm
hello Lufia, This is "The egyptain god" and for the past year i have been a bad person on DMO, i have insulted SSJ4 San Goku many times. He has tried to forgive me, but i ignored him. He has banned me from almost every server on Dmo. I just want you to help. I am really sorry and i apologize to goku for harrassing him. I also apologize to everyone on dmo, i am really sorry. I want to start afresh and i promise not to insult anyone on Dmo. And i am really sorry!
This are all the new keys i made in order to develop a new person. i was orginally, The egyptian gods, then i changed to all these keys, Knightofdestiny, Yugiohmuto, Winner1234, Nowak, ThierryHenry, Dmo Duelist. All i want is for you to help me, and tell goku that i am sorry.
May 7 2007, 10:24 am
i ranked the game,where do i get the member stuff?
Oct 26 2009, 12:31 pm
Oct 26 2009, 3:16 pm
Jan 25 2010, 4:13 am