- Each person gets one vote.
- Each person gets one vote every designated period of time (week, month, year, etc.)
- A game gets a vote based on how many times a link is clicked.
- A game gets a vote based on how many times it's downloaded.
Please tell me which you think is best, or something I missed entirely.
ID:153309
 Jun 18 2004, 6:11 am
|
|
As you may know, I've recently taken it upon myself to rewrite BYOND100 in a fairer format. For example, only games accepted into an official channel will be elligible. But one thing I'm not sure how to do is voting. I've come up with a couple ideas:
| |
|
Jun 18 2004, 6:13 am
|
|
Hazman
 |
I think you should have one vote/day. You could then draw the results at the end of every week, and have an 'overall' winner picked at the end of the month. Many Tot-Game sites use this system, and I think it is quite effective.
|
Yes, so that very popular games didn't always win. If a vastly popular game got 100 votes, and a not so popular one got 50, you'd get a huge gap in them - the popular game's lead would just increase and increase.
What I'm saying is... Clear the votes every month. | |
The weekly totals would just tell you the top game of the week. You could keep a total each week and an all-time total so that you could also see the most popular games ever. And you know the first week a game got a vote, so you could tell the total time its been on the list, so you could have a page show the votes/time so that you can compare new and old games.
| |
Maybe a votes/week average type thing so that games that are relatively new can still compete with older games - game A has been on for 4 weeks with 100 votes, and game B has been on for 1 week and has only 25 votes, but they still have the same vote/week average - 25
| |
Yes, so that very popular games didn't always win. Uh did I miss something :P? Isn't the whole point to give a listing of the most popular games in that order? | |
He meant "so that the games that were popular in the beginning don't always win," or something along those lines. I think.
| |
HavenMaster wrote:
A game gets a vote based on how many times it's downloaded. No, that is a very bad idea. The number of times a game is downloaded has nothing to do with how much those people who downloaded liked it. The best way to go is to give everyone one vote. They can vote as often as they like, but any newer votes cancel out their older votes, so everyone can have their vote supporting a single game all the time. Doing that, you could easily find what the current most popular games are. Then, if you wanted, you could have it keep track of past top games so people can see what games were where on the list at what times (perhaps keep it in a graph form). | |
Better than one vote, perhaps give people several votes, but have them rank their votes. Their first place vote might count for three points; second place might count for two points; and third place would then count for one point. That way people don't have to choose an absolute favourite. =)
| |
If that way was used, something could come in first place when nobody had it as a first place choice. If ten people vote for game A to be first place and 16 people vote for game B to be second place then game B beats game A. If you use multiple votes with points in that way, I suggest you make the margin quite large. Perhaps 10, 5 and 1 points for first second and third, respectively; and maybe some other checks and balances here and there.
| |
Alright everyone, thanks - I have some good ideas for how to handle votes. How do you recommend the list be sorted?
| |
A list of titles currently having the most votes, a list list of titles which have been on the list the longest and perhaps some lists for those which have the most average votes over the last X amount of time.
A graph that shows how many votes any game has had at any point in time would also be nice, but that would require much more work and take up more memory. | |
Loduwijk wrote:
If that way was used, something could come in first place when nobody had it as a first place choice. And? That's kind of half the point. =) If two people put Game X in first place, and four people put Game Y in second place (assuming that nobody else votes for Game X or Game Y here, just to keep things simple) then under my suggestion those two games would have equal ratings. To me, this is a good thing; if a game has four fans, it's more popular than a game that has two fans. However, if those two people like their game twice as much as the four people like their game, then the "amount of liking" kind of balances out. I do agree those margins are a bit narrow, though; I presented them as an example only. On the other hand, ratings of 10, 5, and 1 are a bit high. Perhaps somewhere around 5, 3, and 1 would be a good compromise. | |